Making a scanner with a DSLR [Archive] - Page 2 - Large Format Photography Forum

Brian C. Miller

12-Jan-2012, 08:22

Thanks for the explanation. Do you think that is what Auto Pano Pro is likely doing with their templates in the discussion I linked to above??

I haven't used the software, so I'm writing based on their demonstration video. Some of their stuff went by pretty fast, and there was no audio.

Autopano Pro (http://www.kolor.com/panorama-software-autopano-pro.html) appears to allow very easy interaction with the tiling. When something doesn't work right, you can easily adjust individual tiles, and it looks like there's some quick editing available on individual tiles, too. Nice stuff.

Yes, they do adjustments and edge detection. The video demonstrated a sunset panorama with dark mountains and flat sky. The user then did what looked like a selection of the appropriate areas and the landmark points were highlighted by the software. Then the appropriate stitch was made. I'm guessing that this is what you're referencing as the template mode, and the tiles are placed in a template for user adjustment. The price seems absolutely reasonable, and if I did stitching more than once, I'd definitely buy this software.

What we are doing here is an easy subset of stitching. The target is flat and the camera is perpendicular to the target, so no adjustment need to be made for camera angle (except for possible camera alignment). The camera position is moved in discreet increments in an X-Y axis, and the images follow each other in a grid. The light's color remains the same, so blending is much easier.

Seth: Yes, the GIMP script could be easily run from the command line, no problem. I don't know how easy it would be to integrate complete camera control, etc., through the GIMP, but that's way down the road. Just running a couple of command-line utilities would be sufficient for this project.

There are already at least two panoramic stitching scripts for the GIMP. I think modifications to either one would suffice for this. I can simulate something like this for myself (not the camera control, but the stitching) because my little Canon P&S has a "super" macro mode, and then just build a little X-Y table and move it by hand according to ruler position. Then I'll have a bunch of images for processing.

The only problem I see is using the GIMP in a memory-limited environment. The latest version seems to get a little touchy with a full 4x5 scan from my Epson 3200.

Ben Syverson

12-Jan-2012, 08:57

Brian, you can't get away from edge detection, and you certainly can NOT get away from a soft transition between frames. That's because even at an "ideal" stop like ƒ/5.6 or ƒ/8, there is some amount of vignette, and corner performance is not optimal. Edge detection is necessary unless you want softness or double images in the transition area.

This is not an "easy" subset of stitching. If anything it's a superset, because you'll probably need some custom code in there. I don't have great hope for the GIMP approach.

However, it's doable, and despite Kirk's reservations, dust is likely to be the single feature that makes it all possible. If you're photographing a 24x36mm area at 1:1 with an appropriate amount of overlap (33% would be good), you just need two pieces of dust every 287 square mm. Trust me, your cleanest negative has at least an order of magnitude more than that. Many scanners can't resolve well enough to see fine dust, but this system does.

Mike Anderson

12-Jan-2012, 09:11

I've been following this thread closely so I thought I'd provide a brief summary. Feel free to correct/append.

Goal: create an inexpensive system to scan film that leverages one's digital camera and significantly bests Epson 750. Multiple tiled exposures of the film will be made and stitched together.
Components: (1) Camera support (2) Film holder + light source (3) Custom stitching software or process for using off-the-shelf software.
Camera Support: the camera support is stationary with manual adjustments for distance-to-film and parallel-to-film. Both vertical and horizontal orientations are being pursued. Camera support will accept a wide range of cameras.
Film holder + light source: Camera is stationary so film holder moves. Automated positioning is optional, initial prototyping will use manual positioning of the film holder. Automated positioning can be added on.
Lens: different lens types (SLR macro, enlarger) are being considered.
Focusing: focusing can be done traditionally (looking through the viewfinder) or employ any aid supported by the camera (live view, tethering, etc.).

Kirk Gittings

12-Jan-2012, 09:11

Might the slight vigneting issue be solved by using a crop sensor camera with full frame lenses?

"Brian, you can't get away from edge detection, and you certainly can NOT get away from a soft transition between frames. That's because even at an "ideal" stop like ƒ/5.6 or ƒ/8, there is some amount of vignette, and corner performance is not optimal. Edge detection is necessary unless you want softness or double images in the transition area."

Brian C. Miller

12-Jan-2012, 09:22

Brian, you can't get away from edge detection, and you certainly can NOT get away from a soft transition between frames. That's because even at an "ideal" stop like ƒ/5.6 or ƒ/8, there is some amount of vignette, and corner performance is not optimal. Edge detection is necessary unless you want softness or double images in the transition area.

If there are double images in the transition area, then that area does, in fact, contain landmarks. Once landmarks are found, then landmarks are used. The exact point (once again) is that an area does not have landmarks for positioning. If inadvertent dust serves at the landmark, then all the same, the dust becomes the landmark.

I'm not trying to get away from landmarks. I never said that a soft transition would not be used.

What I have been describing is a system which will work despite a lack of recognized landmarks. This is a fallback method, not a primary method. It is a modification to existing GIMP scripts.

jb7

12-Jan-2012, 10:36

FYI...

It doesn't deal with the specifics of Large Format, or stitching, but it may be of some use-

http://dpbestflow.org/camera/camera-scanning

jm51

12-Jan-2012, 10:43

Could a flatbed scanner be hacked to act as digital paper? A large neg could be done in pieces then stitched.

Failing that, I really like the idea of the camera under the enlarger. Isn't the dof required from an enlarging lens something like 1mm?

Mike Anderson

12-Jan-2012, 10:47

...The camera position is moved in discreet increments in an X-Y axis, and the images follow each other in a grid....

Brian, are you counting on some precision in the x-y positioning of the film (or camera)?

Kirk Gittings

12-Jan-2012, 10:50

The problem AISI with the enlarger scenario as I see it is that a) then you would have to focus with it tethered-not a huge issue and b) more importantly you would have to move the camera to create the tiles rather than the negative. Personally I think it would be easier and more solid to have the camera fixed and move the negative.

jm51

12-Jan-2012, 11:00

Earlier postings mentioned dof in the microns for the moving neg, I don't see that as easy compared to getting a sliding camera table to within 1mm.

Peter J. De Smidt

12-Jan-2012, 11:09

Earlier postings mentioned dof in the microns for the moving neg, I don't see that as easy compared to getting a sliding camera table to within 1mm.

If I remember rightly, the 20 micron range of depth of field is only for higher magnifications than we're looking at here.

Btw, I ordered a sync cord for my SB-28, and it came in the mail just now. It has a threaded PC connector. I wish I would've known about those a long time ago. I absolutely despise non-threaded pc connectors.

Mark Sampson

12-Jan-2012, 11:11

Jumping in late here, but Kodak invented and marketed a scanner c.1991, called the PIW (Photo Imaging Workstation). that essentially had a camera on a stand and used a moving scanner bar beneath the negative. It would do up to 5x7 and some aerial film, and used a Schneider Makro-Componon lens. As well as a Sun SparcStation 20 computer to run it. As this was pre-Photoshop it had its own image-editing software, and would write the finished files to a Photo CD. I scanned thousands of images with one when I worked for Kodak- it did a fine job but I can't now remember its resolution capabilities. That machine lasted until about three years ago when mechanical issues doomed it for good. God only knows what it cost to design, and of course I never knew what EK sold them for (or how many), but the idea could be updated with modern computers and perhaps be affordable.

Brian C. Miller

12-Jan-2012, 11:25

Brian, are you counting on some precision in the x-y positioning of the film (or camera)?

You mean like a stepper motor? Or using an encoder wheel (http://www.societyofrobots.com/sensors_encoder.shtml)?

Yes. Everything has some precision. How precise does it need to be to accomplish a specific task?

Even moving the camera on a plane by hand according to a cheap machinist's ruler would result in a set of images that will have enough accuracy for stitching in an area with no landmarks. The human eye really can distinguish tiny gaps, and that's more than enough for a prototype.

Stitching software is designed for joining images which have an extremely wide variance of sampling accuracy. The commercial software is advertised as being able to construct an image from a hand-held camera. The project goal is a camera on a computer-controlled X-Y axis. The computer moves the camera, triggers the shutter, downloads an image, and repeats until done. After that the images are stitched.

Dave Jeffery

12-Jan-2012, 11:41

Dave, as a fellow canuck I am hinting that you can dig this pit as deep as you like and there ain't no bone in it. The problems are not obvious until you have some montaging experience and that is why I suggest trying one of the commercial scanning systems before investing much time. On the other hand, you can learn lots and have fun if that is the goal.

Thanks! No free lunch as usual.

Mike Anderson

12-Jan-2012, 11:47

...
Yes. Everything has some precision. How precise does it need to be to accomplish a specific task?
...

Imagine an image of a completely uniform sky with a bird in it, the bird is the only landmark. The bird ends up in the middle of a tile.

It sounds like the precision of the placement of the bird in the stitched image depends on the precision of the film/camera placement during capture. True?

(I'm not saying this is inadequate just trying to clarify.)

Peter J. De Smidt

12-Jan-2012, 12:03

I just did a test using my D200 (ISO 100), Nikkor 55mm (F5.6), at 1:1 with a negative illuminated by an SB-28 flash gaff taped to the 6x6cm light mixing box of a Philips pcs2000 enlarger. 1/64th power on the flash gave a very good exposure.

Ben Syverson

12-Jan-2012, 12:26

Brian C. Miller

12-Jan-2012, 12:43

Imagine an image of a completely uniform sky with a bird in it, the bird is the only landmark. The bird ends up in the middle of a tile.

It sounds like the precision of the placement of the bird in the stitched image depends on the precision of the film/camera placement during capture. True?

That's the exact situation that I'm identifying, and I think that's an excellent example. Or how about a moon in a black sky? Same situation.

Using normal stitching software, the bird's location would be indeterminate. There are no landmarks, and the software has no idea of the X-Y grid position in relation to other areas of the whole picture. Using conventional stitching software, the bird could be in any valid capture position. In the corner, over a few, whatever. The software will never be able to figure it out, and human action is necessary.

Using a system where the the number of captures, width and height, are known, and the grid position of the capture is known, it is trivial to simply center the capture in relation to the other captures. Would this precisely match the position compared to a flatbed or drum scanner? Very likely not, but it's good enough for our application.

For instance, a 4x5 image translates to roughly 5 x 4 full-frame 24x36 captures with a bit of overlap. So the bird could be in (1,1), (3,2), or (5,4), and the stitch would still be valid, because the tile's grid location is retained.

Peter J. De Smidt

12-Jan-2012, 13:55

You will find continuous light preferable. Speedlights are the worst for output and color consistency, but even nice studio strobes have variability.

I produced some stop motion animation a while back with strobes, and the result was quite flickery. It wasn't a problem for that project, but for stitching it could result in patchiness.

I'll find out for sure soon enough. Note that the macro folks regularly use flash with focus stacking, sometimes with hundreds of exposures, and their software, e.g. Zerene Stacker or similar, seems to be able to do ok with that. The advantage of flash is the high CRI, 5500Kish white balance, and the minimization of vibration effects due to short flash duration. I'm hoping that the very low power setting leads to decent consistency.

jb7

12-Jan-2012, 14:43

I'd also go with the flash... small variations in exposure should be fixable in the raw converter, or even in the stitching software. Plus, I've figured out a good design for using flash...

I don't do a lot of stitching, but have done some in the past. I use an inexpensive program called DoubleTake, for Mac OS, http://echoone.com/doubletake/ which can handle quite large files. The developer was very helpful when I needed support some years ago, so I sent him an email and he was good enough to respond-

Here's how I outlined the problem-

Hello Henrich-

I don't do much stitching, but when I do, I use your software-

There is a discussion over at the large format forum- http://www.largeformatphotography.info/forum/showthread.php?t=84769

It concerns using a DSLR to photograph large format sheet film, in small tiles, to be stitched together- in effect, to make a DIY scanner, to enable reproduction at high resolution.
The idea might be to make a moving bed to hold the film, which would be moved in quite precise increments to produce a very ordered tile set.

Now, the problem is, that in a single picture there might be quite large areas with no image information - sky, for example-

There are many different stitching software packages, and I don't know them all, I never really needed to-

My enquiry is- might it be possible to adapt your software to take account of the very ordered nature of the tiles?
It might be something as simple as having access to file numbers on screen, as a numerical reference for those blank areas-
it might be the inclusion of guide lines, or it might be that the software asks for the number of rows and columns, to be filled from top left to bottom right-

It's an interesting little problem, for the small number of us still using big film, and I wonder if it might be something you might have a look at?

The thread is quite long, unfortunately, and I think I might have been the one to bring up the stitching issue- most of the later posts voice concern about that-

Thanks for listening, maybe you might consider joining the forum, and addressing some of the issues?

Thanks again-

joseph

His speedy response-

Hi Joseph,

If you always order the images to be stitched exactly the same way, you can use the DoubleTake file format to this.

The doubleTake file format does not contain the images, but just how they are positioned. The images are referred by file name, and if you switch in a different set of files and open a .doubletake file then the new images will be stitched.

Let's say you shoot four images, top-left.jpg, top-right.jpg, bottom-left.jpg and bottom-right.jpg.
Then stitch them with DoubleTake - arranging the images manually.
Save the layout as four-shot.doubletake

Then shoot a new set the exact same way - and replace the 4 images. Open four-shot.doubletake and it will load the new 4 images and you can save the result out.

Kind regards
Henrik

So you see how this might work-

First, draw up a grid, on a sheet of acetate same size as your film, photograph it in position of your film, export the files with custom file names, perhaps numbered, import into DoubleTake, then assemble the sheet in DoubleTake so the grid stitches correctly.

Then go through the operation again, using the film to be scanned. Some accuracy in the positioning would be desirable, and of course, the pictures would need to be taken in the same order as the matrix picture.

Import the images into your raw converter, perform any batch operations necessary, then export with custom file names, the same ones as before, into a folder with a master file name for the master image.

Open the *.doubletake file, which automatically positions the new images in place of the drawn grid image.

I think this approach looks promising; I might be trying it myself.
I have a design for a setup, I might post it in its own thread when I get it built- I think there's a danger of confusion posting it here, and Peter was first to post his in this one anyway- it might be rude to overlap...

Brian C. Miller

12-Jan-2012, 14:57

As long as a computer-controlled X-Y table with a camera can be controlled with a Mac, then the controlling software can simply write the DoubleTake file, no manual intervention needed.

Peter J. De Smidt

12-Jan-2012, 15:00

Good work Joseph! That approach sounds promising.

.
I have a design for a setup, I might post it in its own thread when I get it built- I think there's a danger of confusion posting it here, and Peter was first to post his in this one anyway- it might be rude to overlap...

I don't mind at all either way. Do what you think is best.

jb7

12-Jan-2012, 15:02

Well, there has been some opposition to mechanizing the xy stage- manual positioning would be a cheaper option. Either way, it should work- at least it's a more convincing methodology than I had just a little while ago-

Peter J. De Smidt

12-Jan-2012, 15:14

In a different thread, cardiomac recommended these LED modules for building an enlarger light source:

http://www.cree.com/products/modules_lmr4.asp

They might work really well for our purpose.

Ben Syverson

12-Jan-2012, 18:09

LED would be great. You can either dial in the RGB output of an RGB array of LEDs, or you can gel the output of a white LED.

Essentially, for color negative, you want to output full spectrum blue light in such a way that when the camera is set to "native" white balance (close to daylight), you get a neutral (orange mask renders as gray) image of the negative.

Frank Petronio

12-Jan-2012, 21:49

It has to be vertical because we bought up all of the remaindered 7,369 Bowens Illumintrans to repurpose into this contraption. I'm not saying this but I know Upstairs they're pissed... I bought ad space and am getting the brochures printed next week... So guys, get on it... It doesn't have to work, the software doesn't matter, just mock up a dummy and use iStockPhoto for samples, we need it for the Spring trade shows. 10x10 booth (wait until to you see the bunnies I found...) Presidential Suite at the Hyatt, case of Jamesons!

The promo photos won't show the entire product anyway, we are just using tight crops so nobody can tell how big it is or figure out that it's just a 1970s slide copier with new decals and a translucent plastic decorative panel.

How much is it going to weigh and what do you think we can pack it down to if we let the consumer complete the assembly and alignments? How many Allen nuts should I order? Remember, no Phillips head visible on the product, we are high-tech, it has to be Hex, preferably Metric but we can mix Imperial sizes in if we save at least $0.001 per unit. Hey it's all about margin... make that your mantra, you gotta swim with the sharks to succeed, nothing beats failure like a kick in the balls.

Testing determined that it needs to be purple to project integrity and an out-of-the-box "attitude" so let's see if we can save by using Mylar wraps rather than powder-coating. We'll be shipping them from South Carolina to Xio Zuoug #$%@$ China and back again for the modifications and cosmetic facelift. We can cross ship a few hundred off the books to Turkey for the EU and UK markets since we can knick those bastards for another $100 of margin while avoiding the onarous regulations.

Naming has been hell but the iscan domain just came available again and with Steve Jobs' convenient death, after suitable mourning of course, our consultants have determined that their will be resurgence of nostalgia for late 90s internet names and the whole "i-Mac naming" phenomena. So right now we're batting around "ibscannin-ubstitchin" or the shorter, spunkier version, "ibiatch".

Look, if we all can be team players here, we can get it out to Best Buy in time for Memorial Day, perfect timing. But I should warn you, management is watching this project very closely and the scuttlebutt is that if you can deliver on time, they have a solid buyer in the Eastman Kodak Company, which is pretty damn cool.

If we do sell it to Kodak be sure to polish your resumes because the only other project on our books is the ishredder, but I heard Kodak is looking for a crack team like yourselves - not to dangle too big a carrot too early on!!!

Dave Jeffery

12-Jan-2012, 22:28

Frank,

This guy can sell it easily

http://www.youtube.com/watch?v=Ac7G7xOG2Ag

el french

12-Jan-2012, 22:30

Nearly every image I've shot in the past 4 or 5 years has been a macro panorama. Most have also been focus stacked with thousands of individual images. The biggest difference between what I've been doing and what is required for this project is the difference between orthographic stitching and spherical stitching. Spherical stitching nearly always requires some overlap (10% to 20% or higher for macros) in order to warp the individual frames to fit the projection. The stitching software will usually require landmarks to do accurate stitching. Orthographic stitching on the other hand will not require any overlap if the XY positioning is accurate to 1 pixel. However, for the absolute best results, lens abberations will need to be corrected before stitching, so some overlap will be desireable to enable blending the seams. Good orthographic stitchers allow you to set the XY coordinates for individual frames and/or allow the individual frames to be warped to fit. AutopanoPro and Microsoft ICE do this (I'm not sure about PTAssembler or PTGui and PhotoShop just hasn't worked well for me as a stitcher).

Featureless areas may not have much contrasting detail, but they may have subtle color shifts that are quite visible from frame to frame. Blue sky is one of the hardest things to stitch, not because it's featureless, but because of the color variations. Stitching software can be set to blend the colors between frames, but then you have to ask, is that the same image I would get by using an enlarger? The stitching and focus stacking software can also do a good job of evening out the exposures, but again it's a question of how accurately do you need to reproduce the image. I see quite a bit of variance in the light from my flashes especially when the batteries start to get low, but you won't see it in the final stitched image.

The images of film shown earlier in this thread have proven beyond a reasonably doubt the resolution of a DSLR with a good macro lens is more than adequate for the task. I think the only remaining questions are how can you reliably move the film or camera and how can you ensure even lighting from the first frame to the last and how much tolerance is there is in each of these to get an acceptable final image? Yet another question is how much post-processing do you want to do? I usually inspect each seam at 100% to 200% and manually blend the incongruities. The current crop of stitchers has reduced the manual blending required significantly. The more accurately you can move the film or camera and the more stable your light source is will determine how much post-processing you need to do. Manual movements will probably require more post-processing, but the end result should be very nearly the same.

Another gratuitous sample: http://photosynth.net/view.aspx?cid=3c6a1c2a-7426-41d7-bbea-3f46feefa9c5

SURF

13-Jan-2012, 01:39

Very well said el french.

1."However, for the absolute best results, lens abberations will need to be corrected before stitching."

2."Stitching software can be set to blend the colors between frames, but then you have to ask, is that the same image I would get by using an enlarger?"

3."I think the only remaining questions are how can you reliably move the film or camera and how can you ensure even lighting from the first frame to the last and how much tolerance is there is in each of these to get an acceptable final image?"

I can add that scanner use several calibrations before good scan be possible. Before each scan (!) scanner calibrate the light and certain calibration is needed in scanning with DSLR: actual light and lens geometry profiles are needed to be created somehow.

As for moving parts: it can be easier to move film in X direction and camera in Y direction.

jb7

13-Jan-2012, 05:10

It has to be vertical because we bought up all of the remaindered 7,369 Bowens Illumintrans to repurpose [...]

Oh dear, looks like Frank might have put a deposit on that Gulfstream V already-

There goes the pension plan...

Frank Petronio

13-Jan-2012, 06:37

Let's just hope the sandblasting for the new cosmetic treatment doesn't mess up the optics.

rdenney

13-Jan-2012, 06:44

Let's just hope the sandblasting for the new cosmetic treatment doesn't mess up the optics.

The only word you stupid engineers know is "can't". Sheesh.

Rick "get with the program!" Denney

peter ramm

13-Jan-2012, 07:17

You know, Frank, if you check out my description of the stitched spider above you can probably save a bundle on models and the type of scanner you use wouldn't matter much.

sethlatimer

13-Jan-2012, 07:20

I appreciate everybodies comments, some great ideas. I am new to LF so I don't have a ton of experience, I just ordered a light table over the weekend!

In my head I have been imagining one of the xy stepper tables moving the slide over the light table, with the camera and the table stationary. With that setup each frame shot will have the same piece of light table under it. If our stitching algorithm first took a shot of the light table with no photo on it, it could use that to normalize each photo's exposure for any unevenness. The lens doesn't change so it can correct abberations before the stitch too.

Finally the hardest part is the blue sky. If there is an array of 20 photos of the slide, then there are many different slide orders we can stitch the thing together. The fastest would be to stitch them together in the order they were shot. Now when you use automatic stitching the computer identifies control points, and generates a parameter about the quality of each of the automatically generated control points. This way the computer can distort and bend the photo in smart ways, relying on the best information. If we can alter these algorithms slightly to get all the control points from all the borders and then stitch them in order (highest quality to lowest) then by the time we get to the lowest quality control point borders the algorithm will have more sides to align with.

I am just beginning to digest the panotools source code. I still think this is a manageable project. But I still have no experience with the stepper motor/arduino setup. My plan is to take some 20 shots when my new light table arrives then Ill have data to work with to mock up the merging.
Seth

Jay DeFehr

14-Jan-2012, 15:41

I haven't kept up with this thread very well, so please excuse me if this has already been linked:

http://theagnosticprint.org/future-of-scanning/

Seems relevant.

rdenney

14-Jan-2012, 17:22

I haven't kept up with this thread very well, so please excuse me if this has already been linked:

http://theagnosticprint.org/future-of-scanning/

Seems relevant.

Yes, it was mentioned, and I've read it recently so somebody linked it in this or another thread.

It makes the same assumption Ctein's article makes, which is using a single digital copy photo of an entire negative. To get a good enough copy image for larger formats would require a better DSLR than we own. This article was primarily interested in copying 35mm, where something like a 5D2 is probably fine. Ctein wasn't so specific, but he didn't speculate on the larger formats.

The step we are taking is to use a lesser camera than the 'Blad H4 (one of which I held in my hands today, but it was still the Pentax 645D that made me wet my pants, because I already have a most useful collection of lenses for it), like, say, a Canon 5D, and then move the film under the camera to make a scan in tiles. Each tile would be made 1:1 (or greater). Even at about 15% less magnification, I've already proved to myself that even with modest equipment and a makeshift setup this approach has the potential to significantly better an Epson V750. I think there is plenty of reasonable potential to match a dedicated film scanner.

The article seems correct that the magic will be in the light source. There is concern that flare will put a ceiling on what is possible, given that the light source can't be pushed through an orifice or slit the way it is on a drum or film scanner. But my test was with a highly diffuse light source and the results were usable; much more collimated light sources are easy-peasy, using a condenser head from an enlarger. We'll see.

Tiling also imposes quite an image assembly challenge, especially given the machines we make at home will not likely provide sufficient precision to align based on pixel-accurate registration. Thus, a stitching algorithm that has to match the pattern is likely, and it's still an open question whether existing products will suffice. Again, we'll see.

Where I think the article is correct is that we simply have no choice given what we, as individual photographers, have available to us. We don't represent a sufficient market for a fully integrated machine, and few of use could afford such even if it was made. We will never be able to achieve custom sensing hardware or other specialized products. We might be able to write the software for ourselves, or more likely find ways to work efficiently with what's already out there. But most of us could cobble together a decent DSLR and a quality copy lens--I got my results with a 13MP 5D and an older EL-Nikkor 105 on an adapted bellows.

Rick "hopeful" Denney

Dave Jeffery

14-Jan-2012, 18:14

Jay DeFehr

15-Jan-2012, 17:13

The step we are taking is to use a lesser camera than the 'Blad H4 (one of which I held in my hands today, but it was still the Pentax 645D that made me wet my pants, because I already have a most useful collection of lenses for it), like, say, a Canon 5D, and then move the film under the camera to make a scan in tiles. Each tile would be made 1:1 (or greater). Even at about 15% less magnification, I've already proved to myself that even with modest equipment and a makeshift setup this approach has the potential to significantly better an Epson V750. I think there is plenty of reasonable potential to match a dedicated film scanner.

Where I think the article is correct is that we simply have no choice given what we, as individual photographers, have available to us. We don't represent a sufficient market for a fully integrated machine, and few of use could afford such even if it was made. We will never be able to achieve custom sensing hardware or other specialized products. We might be able to write the software for ourselves, or more likely find ways to work efficiently with what's already out there. But most of us could cobble together a decent DSLR and a quality copy lens--I got my results with a 13MP 5D and an older EL-Nikkor 105 on an adapted bellows.

Rick "hopeful" Denney

I think this is a very interesting project, and I think you guys are doing a great job of identifying the parameters and potential pitfalls, and working through potential solutions. I love these kinds of collaborative efforts.

Incidentally, Tod Gangler has in his studio an old process camera fitted with a MF digital back for copying artwork (reflective). I suppose the platform could work as well for scanning (transmissive) LF and ULF negatives, given a suitable light source.

Good luck!

Greg Blank

15-Jan-2012, 21:24

I worked at a prolab in Baltimore for about a year, they had a digital scanning back for flat art which used something like carbon arc or halogens for exposure. I have often considered getting a FF Dslr so I could rapidly "dupe" some of the twenty-thirty thousand 35mm images I have collecting dust. A capture back would be best rather than a scanning back, that way fluctuation in the source would be eliminated because you could use flash instead of a continous light source.

I think this is a very interesting project, and I think you guys are doing a great job of identifying the parameters and potential pitfalls, and working through potential solutions. I love these kinds of collaborative efforts.

Incidentally, Tod Gangler has in his studio an old process camera fitted with a MF digital back for copying artwork (reflective). I suppose the platform could work as well for scanning (transmissive) LF and ULF negatives, given a suitable light source.

Good luck!

argos33

15-Jan-2012, 21:29

It seems as though creating a template with numbers along the edges would help the stitching software along the sides.

I wonder if very, very small letters or markers had to be placed on the film to help the stitching software if Content Aware Fill would work well enough to repair the scan?

That is a good idea Dave - doing that would surely help with stitching for frame patterns where each frame has one edge along the frame outside of the negative. For example taking 4 or 6 photos of the negative where each smaller frame touches an outside edge. Think rulers in PS, except physically bordering the edges of the negative. As long as you included part of said ruler/guide it would probably help a lot and you could just crop it out of the final stitch. You could probably just make one for each film size, or perhaps an adjustable version like the variable sized printing easels to allow for many different sizes.

Perhaps something like the small letters placed on top of the film could be used as a plan B of sorts, where you could drop one in the sky (where it is having a hard time stitching) and and then easily remove it.

Evan

Jay DeFehr

15-Jan-2012, 23:34

I worked at a prolab in Baltimore for about a year, they had a digital scanning back for flat art which used something like carbon arc or halogens for exposure. I have often considered getting a FF Dslr so I could rapidly "dupe" some of the twenty-thirty thousand 35mm images I have collecting dust. A capture back would be best rather than a scanning back, that way fluctuation in the source would be eliminated because you could use flash instead of a continous light source.

Aren't there devices that attach to the front of a lens for dupe-ing slides? Couldn't one of these devices be adapted for scanning 35mm negs with a DSLR? I wonder how one would compare to a dedicated film scanner? It would certainly be faster than any film scanner I've used.

jb7

16-Jan-2012, 04:42

Is this the world's first flatpack scanner?

Check the new Ikea catalogue, search for Petronio-

jb7

16-Jan-2012, 04:43

Now, if only they had included the instructions...

Jim Michael

16-Jan-2012, 05:15

I worked at a prolab in Baltimore for about a year, they had a digital scanning back for flat art which used something like carbon arc or halogens for exposure.

The Petronio scanner would be scalable to handle large flat art. Replace your Cruse.

Peter J. De Smidt

16-Jan-2012, 15:52

Dr Klaus Schmitt

16-Jan-2012, 16:58

Thanks for the compliment. Wish you luck that it works for you! These Magnogon lenses I was told were used in IMACON scanners sold by Hasselblad. Image circle is 72mm. The only data I ever found is attached FYI.

rdenney

17-Jan-2012, 14:42

I just got a tip from a knowledgeable source about a very good lens for this application. Check out Ebay auction: 220563494846

Apparently, Rodenstock made these lenses for use in scanners, including Imacons. They are optimized for 1:5 to 2:1, which should be perfect. They are a fixed f5.6 lens. The "buy it know" price for these new lenses is about $80 plus $10 shipping each. I ordered one.

I ordered one, too. Whether or not I end up making the scanner, it will be a dandy copy lens for use on my Pentacon Six bellows. The coverage isn't quite enough for 6x6, but it's fine for 645. And being a bit shorter in focal length, it should work at 1:1 in this application without having to add an extension tube to the bellows as I was going to have to do with the longer EL-Nikkor.

With the fixed aperture, I'm assuming that's the aperture the lens is corrected for, and is therefore diffraction-limited at that aperture. The lens looks symmetrical and should therefore be quite free of coma and distortion. If the specs are anything like the APO-Rodagon 75/4, it should be really stunning.

Rick "two left" Denney

Dr Klaus Schmitt

17-Jan-2012, 15:29

I personally think it is a stripped down version of the Apo Rodagon D 75mm in 1x version!
http://www.rodenstock-photo.com/typo3temp/pics/054eee43ae.jpg

1x version: http://www.macrolenses.de/ml_detail.php?ObjektiveNr=154
2x version: http://www.macrolenses.de/ml_detail.php?ObjektiveNr=253

I also have a f4 version of the same lens, so it seems to be common that Rodenstock made special, industrial versions of that one.
http://www.macrolenses.de/ml_detail.php?ObjektiveNr=123

MTF in comparison:
http://farm7.static.flickr.com/6169/6178814563_b481ea7754_b.jpg

Full lens + MTF (also f5.5) data is here (n german): http://www.rodenstock-photo.com/mediabase/original/d_Rodenstock_Vergr_-Obj__43-62__8225.pdf

.

Peter J. De Smidt

17-Jan-2012, 19:24

For the prices that used Rodagon D's are going for, this is a really good deal. Like Rick said, it'll make a fun macro lens even if this scanner thing doesn't work out.

I've been pretty busy with work lately, but I should have some time over the next couple of days to work on the scanner project.

rob

17-Jan-2012, 20:09

When I bought my apo rodagon-N 50mm, it was attached to a scanner CCD board. I have no idea what scanner it was in. I couldn't find any indications of the scanner brand from the board.

argos33

17-Jan-2012, 22:36

Ah blast must have missed it! Let me know how they work out!

jm51

18-Jan-2012, 11:14

This looks cheap enough to hack:

http://www.ebay.co.uk/itm/320824715762?ssPageName=STRK:MEWNX:IT&_trksid=p3984.m1439.l2649

Peter J. De Smidt

30-Jan-2012, 08:41

rdenney

31-Jan-2012, 07:49

(I almost said "dampening" to give Rick something to comment on ;) )

:)

Rick "not seeing any need for watering this setup" Denney

rdenney

31-Jan-2012, 08:59

So while it may not look like I've been doing anything, something I frequently accused of, I've been working on a light source. It should be done in the next few days.

I've also been doing some research. Check out the unit for macro photography in the following build: http://www.photomacrography.net/forum/viewtopic.php?t=11513&postdays=0&postorder=asc&start=0

Personally, I'd be tempted to uses bolts through the granite with nuts and washers instead of epoxying soft brass inserts in the granite, but the epoxied inserts are more elegant.

I would add some damping compound to the aluminum comment. (I almost said "dampening" to give Rick something to comment on ;) )

Remember, these guys are going for 10x, while we are going for 1x. Their vibration concerns are an order of magnitude higher than ours.

I'm rather persuaded by the notion of tight coupling between the camera and the base system. The high-frequency damping and mass of the granite base should take the edge off shutter movement, and I would be concerned about any damped coupling achievable by mortals lacking the necessary avoidance of creep. But these guys are worried about stage movements of one tenth of one sensel, which is 0.6 microns for your 5DII, in order to maintain sensel accuracy at 10x magnification. With my 5D, I can maintain sensel accuracy at 1:1 by restraining the stage to 8 microns (+/- 4 microns). That would correspond to about 63 line pairs/mm (half of the 125 sensels/mm density of the 5D sensor). That's more than good enough for our purposes, it seems to me.

It would be easy for him to brace the top of his column with light rods to the other corners of his granite slab. That would decrease the magnitude of any vibration without inducing lower-frequency resonance.

More generally, I had a discussion with a friend of mine who is a computer graphics expert at a university. He showed me stitched graphics of microscope slides he is making for his wife, who is a research ophthalmologist at the same university. She is using slides made from tissues of retinal disorders, including macular degeneration. His stitches put together terrain that is in the 6mm range, and scales down to the sub-cellular level. High-end stuff indeed.

He suggests that evenness of illumination will be critical, because people can see macro repeating patterns very easily in an image--far more easily than seeing those patterns when they do not repeat. He believes we'll have a hard time weeding out those patterns. He agrees that our use of an enlarger-head light source might help, but he suggests that every component of the system (lens, sensor) will add falloff even with the most even possible illumination. Sure, we can correct that with software systematically, but that may take some effort to do, and it will need to be done before stitching.

He also thinks stitching will be difficult in large, flat expanses of tone. He does not think the grain patterns will provide the necessary detail to provide accurate stitches. But he thinks we are on the right track using software-driven stitch algorithms rather than attempting the mechanical precision necessary to avoid it--that imprecision will cause more artifacts that the stitching will, especially if we overlap more rather than less.

So, I think the next step for me is to experiment with stitching before moving on to big construction. But the light source will be a challenge. If I use an enlarger head, the heat will be a problem unless I get the source out from under the film, and that will require some creativity with a vertical arrangement and the camera looking down onto the film. I don't want to invest too heavily into solving the light source problem until I'm sure of being able to assemble the image.

Rick "waiting on the lens for now" Denney

Peter J. De Smidt

31-Jan-2012, 10:36

I agree that stitching is the big hurdle, and the only way to know will be to try it.

I've got a glass plate for the negative stage, and the test light source is done. Next, I'll build a holder for the glass plate that allows me to register an ABS guide for the negative carrier. Basically it's going to be a rectangular piece of 1/4" black abs sheet. Why? Because it's stable, opaque, and I already have some. I'll cut out a rectangle in the middle of the sheet, and I'll use two dowels to act as registration pins for positioning the guide on the glass plate. (The ABS will have two corresponding holes, and so when the ABS sheet is slid onto the dowels, it should be positioned in a repeatable way on the glass surface.) The cutout will be such that the entire negative can be positioned over the imaging area by sliding the carrier.

Peter J. De Smidt

1-Feb-2012, 12:33

The good news is that the Linos lens arrived today. The bad news is that I don't see any mounting threads.

Brian C. Miller

1-Feb-2012, 13:14

rdenney

1-Feb-2012, 15:17

Map the light source first, then scan the image. Then apply the differential map from the light source to the image, and it's just fine.

For instance, suppose you have a light source that is just flat-out uneven, such as a cheap battery-operated light box. The camera-scanner makes a calibration pass with nothing on the light source, and produces a light map. This is now used as a base reference for the stitched image. Once the stitching is completed the map is applied to the image, resulting in an adjusted image. Now the light source has gone from being problematic to "perfect."

For instance, the light box has fall-off at the edges, and bright spots from the fluorescent lantern tubes behind thin translucent plastic. This is mapped out, and then applied as a correction.

If only the film moves, the map should remain the same for each component image. In that case, you'd want to apply the correction for the component images before stitching. All this is possible, but it's just a bunch of extra steps for those of us not willing to wait for you to write free software for us. :)

I would expect it to be easy enough to write a Photoshop action that would apply the map to each component image as a batch process. That's probably about as far as I would go, but maybe it won't be necessary. I'll test when I get the lens and can piece together a light source, maybe using a D2 head and a mirror.

Of course, some light sources are problematic for time-sensitive reasons, including any fluorescent source. These could be overcome with long exposures (as is the case with cold light heads on enlargers), but it's better to use a non-pulsing light source. Interference in the frequency domain will cause enough problems.

Rick "not wanting to require new software" Denney

Brian C. Miller

1-Feb-2012, 16:31

Rick "not wanting to require new software" Denney

Dear Santa: I want a new product that doesn't need any new hardware or software, and it'll last forever. Sorry about the yummy cookies, the sugar and cholesterol will kill you by summer.

If the solution that someone wants is a manual XY table, then there's nothing stopping them from doing that right now. And I do mean today. If, on the other hand, there's any automation, then custom software must be written, or modified from existing software. My personal preference is for something that works as closely as possible to an Epson, or other, scanner. That's what I'll use, because I don't want to spend a lot of time moving a camera from spot to spot on a table.

-- Brian "Machine, thou art my servant. Slave away for my amusement!" Miller

PS - Look at it this way, at least nobody (yet) is suggesting that chips need to be burned and boards need to be fabbed. I've done both, and it's not difficult.

Peter J. De Smidt

1-Feb-2012, 17:14

http://i955.photobucket.com/albums/ae37/peterdesmidt/Side.jpg
http://i955.photobucket.com/albums/ae37/peterdesmidt/Rear.jpg
http://i955.photobucket.com/albums/ae37/peterdesmidt/front.jpg

There don't appear to be any threads on the back of the lens. On the front, there are female threads, approximately 31mm.

argos33

1-Feb-2012, 18:10

Peter,
What is the outer diameter of the back of the lens? Perhaps you could use something like this:

ebay item #250980129341

Try searching "slip-on lens clamp __mm" or "slip on lens adapter __mm"

Peter J. De Smidt

1-Feb-2012, 18:15

Argos, thanks for the idea. The diameter of the smallest part on the back is 30mm. The main body is 40mm.

rdenney

2-Feb-2012, 07:47

Here's what I will probably attempt (and my lens is waiting at the post office for me to pick it up--tomorrow):

On the back of the board, install three L-brackets, screwed to the board with small screws. The two side of the L are 1.) flat to the board through which those screws go, and 2.) flush with the inside edge of the 30mm hole. Insert the lens into the hole, and these L brackets will rest on the barrel of the lens. Install a radiator hose clamp around the barrel and L brackets, and gently tighten. Cover hose clamp with black tape.

You don't have to tell anyone you did it that way.

Rick "who has used this technique before for lenses with unobtainable flanges" Denney

Peter J. De Smidt

2-Feb-2012, 16:51

That's a good idea, Rick.

I spent some time on my negative stage tonight. I have a 12" square piece of 1/2" thick plate glass. I used to use it to make contact sheets in the darkroom. I checked it with a precision straight edge, and it's not flat. It has about a .5 mm crown. It might be a little less. I don't know if that's good enough or not. To get it really flat, I expect we'd have to have the glass ground and polished, and that's probably expensive.

I wonder if some types of phenolic sheet is naturally flatter? Woodworking shops sell pieces of it to use for flattening the bottom of planes, although people use glass for that as well.

Obviously, if the material isn't clear, a hole would have to be cut for the light source.

8x10 user

2-Feb-2012, 17:52

I still don't see the need for a new scanner. There are many scanners that out resolve large format film. High end drum and flatbed scanners are able to produce scans with zero or close to zero added noise. I could see using a grid rather then linear sensor when you want more speed but a linear sensor is good because of a lack of bayer issues. There a few work around's to get pure RGB data from a grid based sensor but....

Peter J. De Smidt

2-Feb-2012, 18:24

I don't see the need to argue about it. The justifications for making one have already been stated in the thread. You don't have to make one.

jb7

3-Feb-2012, 02:52

He suggests that evenness of illumination will be critical, because people can see macro repeating patterns very easily in an image--far more easily than seeing those patterns when they do not repeat. He believes we'll have a hard time weeding out those patterns. He agrees that our use of an enlarger-head light source might help, but he suggests that every component of the system (lens, sensor) will add falloff even with the most even possible illumination. Sure, we can correct that with software systematically, but that may take some effort to do, and it will need to be done before stitching.

He also thinks stitching will be difficult in large, flat expanses of tone. He does not think the grain patterns will provide the necessary detail to provide accurate stitches. But he thinks we are on the right track using software-driven stitch algorithms rather than attempting the mechanical precision necessary to avoid it--that imprecision will cause more artifacts that the stitching will, especially if we overlap more rather than less.

I've built the platform, and have done some initial testing. I've also got a much improved design for a Mk II, but until it can be proven that the combination of uneven illumination (light source and lens) can be corrected, I doubt if I will build it.

Regarding the bed; I'm using a 10mm porcelain tile, very flat. I have a piece of 6mm glass that is positioned on top of it, with the transparency taped to that. The glass is guided using some Mod 1 racks, these allow incremental repositioning in multiples of 3mm. Wet mounting might be better, particularly with film that has been clipped in frames; the clips and the heat of the process tends to distort colour films.

I'm going to do at least one more run of tests, but I don't hold out too much hope of making an acceptable scan, for an acceptable level of commitment...

The magnification of the image that's possible is quite scary, right into the dye clouds-but not as scary as the magnification of the dust...

Peter J. De Smidt

3-Feb-2012, 06:05

Joseph,

Thank you very much for building a prototype.

Do you have any pictures?

rdenney

3-Feb-2012, 06:34

I still don't see the need for a new scanner. There are many scanners that out resolve large format film. High end drum and flatbed scanners are able to produce scans with zero or close to zero added noise.

It just depends on how much money you have. When those high-end scanner mechanical parts wear out and can't be replaced, can you afford to fabricate replacements? Do you have the space for the sort of scanner you describe, plus the computer museum you will need, say, 10 years from now? What will you do if the circuit board fails? What if you can't afford over a hundred bucks per scan to those willing to maintain the obsolete technology?

Better to consider alternatives now, before they become an absolute necessity, and for those for whom the scanners you describe are impractical or infeasible, especially considering they might have a life's work in negatives that they hope to mine in future decades.

Rick "glad early photographers were a bit more experimental" Denney

Kirk Gittings

3-Feb-2012, 08:15

In addition to what Rick said above, if this works out, as time passes and scanners are going away we will have ever better dslrs to make this solution continually improve.

jb7

3-Feb-2012, 08:19

Thanks Peter- no, I haven't photographed it yet, I will when I have it set up for testing again. Not sure when that will be, I'm busy on other things at the moment.

Stitching is another thorny issue, quite apart from the repeating vignette problem.

I made the scanner bed oversize- since I may have need to scan larger images in the future, larger than 8x10. A smaller size, the MkII, would have better ergonomics.

First things first though, need to improve the light source...

Nathan Potter

3-Feb-2012, 08:38

Rick said it nicely. Many of us have fine DSLRs already paid for along with a fine macro lens. The image capture and signal processing within the DSLR is phenominal, so worth making use of. Setting up to copy film by backlighting is not such a big deal, depending on how fancy you want to be. The stitching has to be worked out and is manually time consuming but the high potential quality of scan would be limited to only those images that deserve it, so it would be used sparingly.

Any 6um pixel DSLR used at 1:1 with a macro lens compatible with that resolution will yield a digital frame at 4200 spi and at an astoundingly high quality, especially with a reasonably collimated light source. That would easily match what you can get from a Nikon 5000 ED type device. We're talking home made lashups here not engineering a commercial product.

Nate Potter, Austin TX.

Peter J. De Smidt

3-Feb-2012, 09:06

Here's my Light Source Prototype 1 (LSP-1).

http://i955.photobucket.com/albums/ae37/peterdesmidt/LSP1_NoLight.jpg

http://i955.photobucket.com/albums/ae37/peterdesmidt/LSP1_LED.jpg

http://i955.photobucket.com/albums/ae37/peterdesmidt/LSP1_Flash.jpg

It's built like the inside of an enlarger light box, i.e. it's lined with white styrofoam, and the side opposite the light source is at a 45* angle.

Peter J. De Smidt

4-Feb-2012, 08:09

Asher Kelman

4-Feb-2012, 10:02

Well it seems older drum scanner used photomultiplier tubes. I'm betting you can't get those anymore. Newer ones use CCD arrays which are not easy for amateurs to deal with. It seems more practical to put a whole bunch of APS-C sensors on a device that can position them. You take one picture with the whole bunch of sensors, shift the sensors (because there will be borders between the sensors), take the picture again and stitch the result. If this could be done with an 8x10 cameras I'm sure the resulting digital image would be in the Gigapixel range easily.

Two lines of sensors with the second one covering the gaps in the first. That would then scan over a flat pattern. Nothing complicated.

Asher

argos33

4-Feb-2012, 10:14

I did a quick test by photographing the light source with a 55mm Nikkor at 1:1. The middle readings were all L (luminance channel LAB) 71. One far edge dropped to L 69, and the other edge dropped to 67. I can tray and position the flash a little better, and if that doesn't work, I can add a fresnel to or more diffusion in front of the flash, but I'm certainly open to suggestions.

Peter,
A few things I thought of looking at the prototype:

1. The diffusion plane/light source doesn't necessarily have to be where the negative sits. You can use any materials to get a good diffused light source, and use a flat piece of glass, etc for the negative to sit on above that. Perhaps you can use a thicker and/or more opaque piece of plexiglass or acrylic sheet for the top panel?

2. If there is just a little falloff on the edges perhaps you can just mask the surface so that the light source area is smaller?

3. Could you scoot the flash back a little bit so it's cone of light is wider? I assume you have the flash set on the "wide" setting?

Looks good though, that's a good idea adding the 45 degree bounce, it could save a lot of vertical room that way.

Evan

Frank Petronio

4-Feb-2012, 10:51

Geez I ignore "my" thread for a couple of weeks and you guys are working away but when are we doing the naming and brand identity?

Cracking the whip... marketing comes first.

Peter J. De Smidt

4-Feb-2012, 11:08

Evan, thanks for the suggestions!

Peter,
A few things I thought of looking at the prototype:

1. The diffusion plane/light source doesn't necessarily have to be where the negative sits. You can use any materials to get a good diffused light source, and use a flat piece of glass, etc for the negative to sit on above that. Perhaps you can use a thicker and/or more opaque piece of plexiglass or acrylic sheet for the top panel?

Yep, that's the idea. The light source will sit under the glass plate on which the negative carrier will slide.

rdenney

4-Feb-2012, 11:09

Peter J. De Smidt

4-Feb-2012, 11:09

Geez I ignore "my" thread for a couple of weeks and you guys are working away but when are we doing the naming and brand identity?

Cracking the whip... marketing comes first.

Well, we can call it the "Petronio Modular Scanner", or PMS for short ;)

roresteen

4-Feb-2012, 14:50

Well, we can call it the "Petronio Modular Scanner", or PMS for short ;)

I'm for that..Frank, great idea here...I really don't see why it can't be done...I love your open source idea...it shouldn't have to cost that much.

Frank Petronio

5-Feb-2012, 06:56

Peter, you could have had a nice career at http://www.siegelgale.com/....

(They are the people who can BS places like K-Mart into spending $6M for a large red "K", Xerox into a large red "X", etc.)

Peter J. De Smidt

5-Feb-2012, 07:34

Or how about DIYS, or Do It Yourself Scanner. Let's pronounce it like "DICE", since making one of these is a bit of a gamble. If we're not careful, we'll lose the 'Y' and end up in Dis, the city that contains the lower levels of hell.

I nominate Frank for spearheading our marketing campaign, and in particular for finding our attractive female spokes-model. Obviously, we'll need a youtube video, perhaps something along the lines of the opening of 2001, complete with an Also Sprach Zarathustra soundtrack. That's a nice subtle implication that those who use the scanner would immediately qualify as Nietzschean supermen. The negative being scanned should, of course, be one of our spokes-model.

Frank Petronio

5-Feb-2012, 07:45

We could throw a scanning drum through an Epson platen ala Apple's "1984" commercial.

Jim Michael

5-Feb-2012, 07:53

You're already very close. I don't think I'd do more with the light source until you've finalized the lens selection. The lens will add its own falloff characteristics.

Then, I wonder if you could print a reflector with a counter pattern for that 45-degree panel. It would be subsequently diffused and defocused and a print on an Epson on matte paper could work fine.

Rick "loving the simplicity of this arrangement" Denney

In researching some stuff for a 3-D cinematography project I learned that custom sized first surface mirrors are cheap. So the 45 degree reflector styro could be replaced with that and add the counter-pattern if required.

I don't think anyone has tried the collimated source yet. I have a condenser from my D-2 I may try out.

Also was thinking about my manual grid system. Originally was thinking of using perforated sheet metal and have a carrier with pins to fit in the holes. A more flexible approach might use a printed paper grid under glass and the carrier just slides around on the glass. When you want a different spacing just switch out the grid paper. Grids are pretty easy to create in postscript.

marfa boomboom tx

5-Feb-2012, 09:13

may provide bits of concept... very old scanner conversion for different purpose, but he did have to solve several of the same problems.

http://groups.csail.mit.edu/graphics/pubs/thesis_jcyang.pdf
"The primary motivations for this project is to reduce the acquisition cost of light-field capture devices and create a portable system suitable for acquiring outdoor scenes. This paper describes the design of such an apparatus using readily available parts. One of the strategies for reducing the system cost has been to rely on software to correct as many of the geometric and photometric inaccuracies as possible. The resulting light field acquisition device can be built for under $200.
The presented light-field acquisition system employs a modified low- cost flatbed scanner. The scanner is interfaced to a standard desktop PC for indoor use or a laptop for outdoor experiments. Focused onto the glass of the scanner is an 8-by-11-grid assembly of one-inch plastic lenses. To make the scanner mobile, the DC power supply is replaced with a 12V lead acid battery. Although the construction of our light-field capture system is simple, the most significant challenges involve the processing of the raw scanner output. Necessary adjustments are color correction and radial distortion removal."

use software whenever you can....

Ben Syverson

5-Feb-2012, 10:12

We could throw a scanning drum through an Epson platen ala Apple's "1984" commercial.
On December 21, LFF will introduce DIYS, and you'll see why 2012 won't be like 2012.

Peter J. De Smidt

5-Feb-2012, 10:54

may provide bits of concept...

Thanks for the info!

That's a much more useful Master's Thesis than I did. Mine was titled, "Anti-Realism and Scepticism: Crispin Wright on Imploding Descartes' Demon."

rdenney

5-Feb-2012, 21:44

Thanks for the info!

That's a much more useful Master's Thesis than I did. Mine was titled, "Anti-Realism and Scepticism: Crispin Wright on Imploding Descartes' Demon."

You think that was useless? Mine was "Traffic Platoon Dispersion Modeling, Evaluating the Mechanism and the Empirical Support."

Back to topic: I'm fiddling with the Linos lens. I'm thinking, based on dimensions, this is a Rodagan D-2x, not the 1x, and is thus optimized at 1:2 rather than 1:1. The 1:1 version was about 8mm longer occording the Rodenstock's catalog. Diameter's bigger, too. And thinking about the CCD size in an Imacon, it would need to be optimized for less magnification than 1:1.

I'm just not feeling it with this lens. Maybe I just can't get it focused accurately (probable) but I'm just not seeing quite the resolution I saw with the Nikkor enlarging lens. I'm holding the lens up to a flat mounting surface, so maybe that's an issue, too. I need to mount it more carefully and devise a better focusing arrangement--one that will also really lock down the camera. But I'm pretty confident camera shake isn't an issue.

Have you tested it yet? My sensor pixel pitch might not be high enough, and the lack of live view might be really debilitating. The lens field seems flat enough and I'm not detecting any distortion, so maybe I just need to focus it better. At it's fixed f/5.6, focus is really critical compared to the f/11 I was using with the Nikkor.

Does the Imacon focus at 1:1 for roll film (I know it does not for 4x5)? It could--it's something like 3200 spi which is about 125 sensels/mm at 1:1.

Rick "who'll set this aside until you tell me it's worth the trouble to mount it" Denney

Frank Petronio

6-Feb-2012, 03:22

Rick, I'd like green lights on my way into town today.

Peter J. De Smidt

6-Feb-2012, 04:51

No, I haven't played with the lens yet.

On the other hand, I made progress on the negative stage yesterday. Assuming I have a little bit of time today, I should get it completed. I'd really like to see how samples at 1:1 will stitch. I'm planning on using a 6x7 negative, and either my Nikkor 55mm or a Rodagon 80mm, not because they will be the ultimate in quality, but because they're ready to go.

marfa boomboom tx

6-Feb-2012, 07:24

along the lines of, do more with software, this is a link to imagestack

http://code.google.com/p/imagestack/

which is used at Stanford in their frankenCamera course.

http://news.stanford.edu/news/2009/august31/levoy-opensource-camera-090109.html

Recall, that Leaf began as a scanner : http://thedustylenscap.com/2012/02/04/from-science-to-art-the-story-of-leaf/

Digitizing is the same. What you choose makes the difference. Software doesn't know what the hell it's working on, only you do.

rdenney

6-Feb-2012, 09:27

Rick, I'd like green lights on my way into town today.

So would I.

Rick "who'll leave it at that" Denney

marfa boomboom tx

7-Feb-2012, 09:02

found off googlecode ..

' How to take a macro gigapan with a few pieces of wood

I like complicated things but you can explore the macro world with a few pieces of wood and a little bit of patience. Here is the 'simplest least useless thing' for Macro imaging "
from: http://www.explorablemicroscopy.org/

These folks are making scans using very cheap materials, scrap wood etc. Their results are achieved using software from: http://code.google.com/p/nanogigapan/source/checkout

My summary: many are doing or have made cheap step and repeat photo systems. The hardware seems to be available from robotics and DIY electronic sites. The software is available as source for most of the pieces. What is missing is the "assembly"

---

Peter J. De Smidt

7-Feb-2012, 10:12

That looks like great info. Thanks!

federico9001

7-Feb-2012, 15:55

The best drum scanners (in perfect conditions) are in another league than any CCD scanner, professional or not.
Then the point is the operator's skills (to create excellent profiles-calibration, to achieve a perfect workflow etc.).

I do drum scans for international artists and photographers.

Take a look:
http://www.flickr.com/photos/castorscan/6766682735/in/photostream/

http://www.flickr.com/photos/castorscan

Peter J. De Smidt

7-Feb-2012, 16:19

And that is a positive contribution to this thread because...?

Yes, drum scanners are the best quality scanners currently available, and there are a number of fine drum scanner operators who take part in these forums. The reasons we are engaged in our project have been well-detailed in this thread. If you are interested in making a positive contribution to the project, then you are welcome to participate. If you're only interested in hawking your wares, then please do so elsewhere.

Peter J. De Smidt

7-Feb-2012, 16:23

I finished the negative stage today. I'll have to make some changes to the camera support structure before running any tests, and I'm trying to decide whether to use my 80mm Rodagon on a bellows or my Nikkor 55mm micro. Neither will be ideal, but I'd like to get to trying to stitch the frames fairly quickly, as that appears to be the biggest hurdle facing us.

On another note, the D800 announcement bodes well for the future. While I certainly won't be able to afford one, or something similar, for many years, this type of resolution could really cut down on the number of samples required for a scan. I expect that for most 35mm, no stitching would be needed. This would open up more time for hdr or focus stacking and still the process should be much faster than any current scanner.

federico9001

7-Feb-2012, 17:10

Hi Peter, how you doing?

I'm marco, I contacted you approximately 1 year ago about the Screen Cezanne...do you remember?

8x10 user

7-Feb-2012, 20:26

I still think there are better alternatives but if your going to DIY you should try using a black and white CCD sensor and RGB filters on a color wheel. Also you should be able to set the microns of your sample and the sample frequency independently.

If you sample a space smaller then your dye clouds then you get crazy color noise or "grain aliasing"... Its when you see the little red green and blue parts rather then what they look like as a whole.

The sample frequency or "resolution" can be increased past the sample size and will continue to improve the image. This is due to the stochastic nature of film.

rdenney

7-Feb-2012, 21:10

I still think there are better alternatives but if your going to DIY you should try using a black and white CCD sensor and RGB filters on a color wheel. Also you should be able to set the microns of your sample and the sample frequency independently.

Where does one find a black and white CCD sensor in packaging that can be replaced to overcome obsolescence in the future?

Rick "who could design all sorts of stuff that would fail in the market just as current scanners are doing" Denney

Peter J. De Smidt

7-Feb-2012, 21:18

peter ramm

8-Feb-2012, 07:57

Well, if you have a black and white ccd sensor that you'd like to lend me, then I'd be happy to pursue that route.

I do have a dslr, old as it is. And the tests run so far indicate that the quality minimum we are after is available in a single capture of a small area of a frame of film.

Peter, as I understand it the quality minimum you are after is defined by the performance of a typical consumer flat bed scanner, such as the Epson. Let me try to be helpful in telling you what you will find.

On a single acquisition

1. The advantage of the area sensor (DSLR with an X x Y chip) over a linear sensor (Epson) is that you can magnify as much as you like in a device that images quickly. Very useful in imaging fine details of small targets while retaining MTF. However, larger targets (e.g. 4 x 5 film) require so many acquisitions that they become impractical. This is with a motor stage or rail system to move the sensor or target. Manual movement will fail in actual use, both because angular errors corrupt the alignment algorithms (your hand tilts the film a bit), and because it is just so tedious.

2. An area sensor will lose intensity resolution (precision, tonality, whatever you want to call it) in both dark and light areas. Your shadows will block up, and subtle intensity transitions in the light areas will be lost. These issues are generated both in software (correcting spatial intensity variation) and in the optical chain (flare). They are not solvable without altering the acquisition mode to one that is flare resistant (spot or line scanning). In other words, spatial resolution is the least of your problems. We shoot film for the tonality more than the detail and that is why scanners exist.

3. Color makes the problems much worse.

Stitching

1. Monochrome stitching suffers from an interaction between the correction field (your reference) and the intensities in the film. In other words, overall light levels during the scan affect the spatial distribution of the correction field. That results in noticeable intensity variation in the composite image (patches), with alignment errors also contributing to patchiness. Careful manual control/editing of the process helps, but it can take hours to do something like a 10 x 10 montage at high quality. Again, this is with a motor stage.
2. Color stitiching is a fearsomely difficult process for lots of reasons. Color images will be patchwork and muddy (because of loss of intensity resolution) unless you have discrete chips for each primary or a filter wheel. Even then, there will be a loss of color rendition and a subtle patch effect. Not better than Epson.

I will now shut up.

Ben Syverson

8-Feb-2012, 07:58

Keep on rockin, Peter! Don't pay attention to the naysayers. Especially the ones with lots of technical suggestions but no actual technical expertise.

Ben Syverson

8-Feb-2012, 08:17

(snip) However, larger targets (e.g. 4 x 5 film) require so many acquisitions that they become impractical. This is with a motor stage or rail system to move the sensor or target.
It sounds like you're speaking from experience. Are you? Because doing four rows of six exposures (24 total shots) does not sound "impractical" to me.

2. An area sensor will lose intensity resolution (precision, tonality, whatever you want to call it) in both dark and light areas. Your shadows will block up, and subtle intensity transitions in the light areas will be lost. These issues are generated both in software (correcting spatial intensity variation) and in the optical chain (flare). They are not solvable without altering the acquisition mode to one that is flare resistant (spot or line scanning). In other words, spatial resolution is the least of your problems. We shoot film for the tonality more than the detail and that is why scanners exist.
You may be shocked to learn that there is a gigantic amount of flare in a flatbed scanner. There is flare from the scanner platen, as well as from the cheap plastic lens in front of the linear CCD.

In contrast, a well-corrected macro lens at ƒ8 is incredibly flare resistant. I've verified this experimentally, but it could easily be guessed by the superiority of modern glass optics over the plastic scanner optics. Negatives are relatively low contrast, so you just don't excite much flaring.

3. Color makes the problems much worse.

Why? No explanation.

1. Monochrome stitching suffers from an interaction between the correction field (your reference) and the intensities in the film.
I think you're referring to vignette, but it's not entirely clear. Vignette is easily removed mathematically by shooting a reference frame at the target distance and aperture. There is no technical reason why 2-3 dozen very carefully shot frames cannot be aligned and stitched with the proper software.

2. Color stitiching is a fearsomely difficult process for lots of reasons. Color images will be patchwork and muddy (because of loss of intensity resolution) unless you have discrete chips for each primary or a filter wheel. Even then, there will be a loss of color rendition and a subtle patch effect. Not better than Epson.
I'm not sure what "loss of intensity resolution" is supposed to refer to, because you actually gain luminance resolution in the proposed system. "Patching" happens with vignette, which I've already addressed. You don't need a filter wheel or monochrome sensor. There's more than enough color sensitivity in the average DSLR at low ISO.

I've shot my own tests, imaging color negatives at 1:1 with a 5D II, and saw immediately that this method produces sharper, clearer images than a flatbed. It sounds like you speak from experience, in which case I'd love to hear about it. Because virtually every point you make is contrary to my real-world findings.

Peter J. De Smidt

8-Feb-2012, 08:38

I expect that Mr. Ramm knows very well what he is talking about, and this project may very well fail. I'm ok with that, but I'm not going to stop without giving it a try, especially since I know someone who has built such as system for a client, and the client is very happy with the results. And even if they results currently obtained aren't that good, there's been some major advances in all areas the project touches on, digital sensors, robotics, software development..., and I expect these advances will continue. Thus, what might not be practical now might very well become so over the next couple of years.

peter ramm

8-Feb-2012, 08:53

Ben, I have designed camera-based scanning systems and spent infinite hours using them at everything from macro to micro levels. It is not my intention to be a naysayer, or to get into detail about sources of flare in linear vs area scanners. I encourage you to experiment or to buy a beer for an optical engineer. I just believe that using a DSLR is not the way if the end result must be better than an Epson. This is particularly true with color.

You ask for an explanation of the color problems and I avoided that because there are so many. Here is just one. The spatial distribution of response inhomogeneity (more than vignetting) is discrete to each primary, and is both temperature (sensor) and intensity dependent. We were never able to create perfect color correction algorithms for cameras and the eye is very sensitive to that.

Perhaps I should have shut up before the last post.

Ben Syverson

8-Feb-2012, 09:17

Peter, I appreciate your feedback, especially now that I have the knowledge that you've put a lot of time into the problem.

I specialize in image processing, so I'd be curious to hear more about "spatial distribution of response inhomogeneity." You're implying that the color response of a camera CCD varies across the frame? I have not found that to be the case. My friend Justin and I created a wide angle pinhole for Micro 4/3 cameras which does indeed cause localized color shifts due to the incident light angle at the sensor. But such a thing would not happen with a mostly telecentric macro lens, and besides, it's trivial to correct:

http://bensyverson.com/images/2012/02/field-input.jpg

http://bensyverson.com/images/2012/02/field-white.jpg

http://bensyverson.com/images/2012/02/field-corrected.jpg

In the above example, I've chosen to remove only the color shifts, and not the vignette. But once you have a correction frame, you can remove localized color shift, vignette and dust spots in one pass.

Peter J. De Smidt

8-Feb-2012, 09:34

I'm going to take a break from looking at this thread until I have "scanned" a 6x7 frame with my system, using a lens at 1:1. It may be a few days until I can accomplish this.

rdenney

8-Feb-2012, 09:52

My (very slow) progress on this front has been: I have ordered a reversing ring for the Pentacon Six mount, which will go into my Pentacon Six bellows. It provides a 62mm male filter thread. I have also order three sets of "filter stack caps", which provide a solid metal mail and female filter cover to protect the ends of a stack of filters. The female one becomes a lens board that will screw into the reversing ring after mounting the lens in it. That should provide better alignment than using plastic body caps, heh.

I will then mount the Nikkor and the LiNOS lenses, and continue my tests.

I've also looked at the Manfrotto precision focus track, which seems a reasonable enough mid-price focus track. I may end up with three--two for X-Y movement of a negative stage and one for focusing the camera with the bellows fixed to 1:whatever. That track could be very firmly mounted.

It occurs to me that if the LiNOS lens is optimized for 1:2 instead of 1:1, it could be reversed (which Rodenstock claims is acceptable). I think I'll mount it in way that allows reversal--some sort of friction collar on the outer barrel.

Waiting for the arrival of those bits should give you a chance to take another step or two.

Rick "plodding along behind you" Denney

Jim Michael

8-Feb-2012, 10:38

Nathan Potter

8-Feb-2012, 11:14

I think I see Peter Ramms' comments as understandably valid issues but would question the applicability to what is being attempted here.

A lot of people do stitching of digitally captured frames and seem to overcome the issues with frame edge color and image distortions to end up with seamless final images. Clearly one would use a highly rectilinear macro lens with virtually no edge distortion along with apparatus to insure ultimate planarity between copy and camera sensor.

Certainly the DSLR sensors are sensitive to temperature and light intensity (carrier density in the sensor active device is a function of temperature and light intensity) but even using the DSLR out in the field seems not to negate the ability to reconstruct a multi frame final image after the fact.

Hundreds of frames stitched from a Gigapan setup seem to have overcome these issues mentioned by Peter unless I'm missing something here.

Meanwhile, you characters are getting me interested in fooling around with this project.

Nate Potter, Austin TX.

Kirk Gittings

8-Feb-2012, 11:28

Back in the scanner comparison thread there were comparisons based on the same target. Is that target still available? It would be useful to compare the results from that comparison thread to results using this methodology.

I seem to remember that Ted was in the possession of that at the time of his death. I doubt it can be found at this point.

marfa boomboom tx

8-Feb-2012, 11:52

the path I'm pursuing [links to items in early posts in this thread] is sketched:

// I think the attachment is making it, since I see it in 'preview' //

Joshua Dunn

8-Feb-2012, 12:20

I've also looked at the Manfrotto precision focus track, which seems a reasonable enough mid-price focus track. I may end up with three--two for X-Y movement of a negative stage and one for focusing the camera with the bellows fixed to 1:whatever. That track could be very firmly mounted.

Rick,

If the “Manfrotto precision focus track” you mentioned is the Manfrotto 454 Micrometric Positioning Sliding Plate (http://www.bhphotovideo.com/c/product/554343-REG/Manfrotto_454_454_Micrometric_Positioning_Sliding.html)I have one you can borrow if it will help.

-Joshua

rdenney

8-Feb-2012, 12:55

Rick,

If the “Manfrotto precision focus track” you mentioned is the Manfrotto 454 Micrometric Positioning Sliding Plate (http://www.bhphotovideo.com/c/product/554343-REG/Manfrotto_454_454_Micrometric_Positioning_Sliding.html)I have one you can borrow if it will help.

-Joshua

Yes, that's the one. I'm moving too slowly and fitfully on this to be willing to tie up someone else's stuff. I don't want to borrow something and then stall out because other responsibilities emerge. But you can help me with what I can't tell from pictures.

1.) How sturdy is it? When locked down, will it hold a DSLR camera very (I mean, very) securely?

2.) I understand that the track wiggles back and forth a bit when adjusting. This is not a concern for this application, I don't think. But does it bob up and down?

3.) What is the range of motion? The specs were not clear, but I gathered it was 120mm.

It looks to me much sturdier than the cheapie Chinese knockoffs on ebay. I saw one of those at the local store and found it to be rather lightweight for the task. But it's not nearly as expensive as the similar stuff from Really Right Stuff or Novoflex.

Rick "curious" Denney

Joshua Dunn

9-Feb-2012, 00:47

1.) How sturdy is it? When locked down, will it hold a DSLR camera very (I mean, very) securely?

Overall it’s a very sturdy device. As far as how well a DSLR will clamp down on it that will probably depend on which DSLR. But a fair comparison is that it attaches a camera as well as any of the Bogen tripod heads with a quick release plate.

2.) I understand that the track wiggles back and forth a bit when adjusting. This is not a concern for this application, I don't think. But does it bob up and down?

As far as “Bobbing and Wiggling” the device does not have much play at all. The brass knobs on either end of the plate are to adjust the movement of the plate. It is a worm drive, very slow but methodical and very little play. The brass knob on the side locks the drive down solid. I think there is more potential for the camera to move on the plate than any play in the device to throw things out of alignment.

3.) What is the range of motion? The specs were not clear, but I gathered it was 120mm.

The range of motion is 120mm as you surmised.

The only other thing I might add is the bottom of the plate where you would normally attach your tripod (not so using your application) has both 1/4” and 3/8” tripod threads. This could be important for your design if you are using two of these for your X and Y axis. For this to work the X and Y axis would have to be at right angles to each other, any deviation could cause things to go out of alignment. Having two threaded tripod sockets (as opposed to just one) would allow you to bolt them down to each other in a manner that they could not rotate out of position.

I hope this helps. If you get to a point that you might want to use it just PM me and we’ll find a way to get it to you.

-Joshua

rdenney

9-Feb-2012, 05:57

Overall it’s a very sturdy device.

Thanks. I'm considering my options. I've had pointed out to me an interesting device used by precision machinists which is priced similarly and might work better for focusing the camera. But two of these are right angles are still in the running for film-stage motion, even though the worm drive will probably slow that down a bit. The 120mm range of motion is useful, though, for 4x5 film.

Rick "moving very slowly on this project" Denney

marfa boomboom tx

10-Feb-2012, 09:07

just another sample...

http://www.youtube.com/watch?v=If1QeTXImzY&feature=related

Kirk Gittings

10-Feb-2012, 09:36

Does anyone have a problem with renaming this thread something like "making a scanner with a dslr"? As it stands this considerable thread is not about making a new drum scanner and searches would be better targeted if I changed the name.

jb7

10-Feb-2012, 11:15

rdenney

10-Feb-2012, 11:54

How about 'Damn you, Frank Petronio' ?

Speaking of an unspecified search parameter...

Rick "who thought the objective was to narrow down searches" Denney

Nathan Potter

11-Feb-2012, 18:41

At Peters suggestion here is a concept drawing for a DSLR scanner that I posted in another thread. Clearly this is where there is activity on this, so here it is. No big deal, but this is how I might approach this in a vertical configuration.

To maximize contrast I suspect that one needs to use a collimated light source. Drum scanners essentially do that by applying the transmitted light through a narrow angle as seen by a detector, typically a lens and thence to a PM tube. For the ultimate in S/N ratio they may use an integrating sphere with a PM attached, although I don't know if any actually use that technology - possibly unnecessary or awkward in the scheme of things.

In my sketch it may be desirable to move the plano/convex condensers up under the X/Y stage for a sort of upside down D2 configuration. I've assumed about 75 mm diameter condenser lenses (Edmund Scientific @ $75 ea.). The point source might be on the order of 6 to 10 inches from the condensers. I've used this arrangement for other purposes with considerable success (it drastically reduces light scattering in the emulsion).

The support system needs to have the stability of an industrial quality microscope and of course exact planarity between stage and sensor is a must since the DSLR wants to be utilized at near maximum aperture. Obtaining a stage with a 4X5 minimum window is a heck of a problem but I was thinking of that from say a Nikon comparitor as used in machine shops - say a junk piece.

This is really a pretty simple lash up if the parts can be obtained. The ultimate problem will be with the collimated light source and the precise imaging of defects and dust that will be a plague. While of no concern for experiments and proof of concept the final version will need a sealed optical path with access for an occasional cleaning.

With luck herewith is a drawing:

http://farm8.staticflickr.com/7045/6851712881_d94fbc3ebd_z.jpg (http://www.flickr.com/photos/argiolus/6851712881/)
SCANNER-3jpg (http://www.flickr.com/photos/argiolus/6851712881/) by hypolimnas (http://www.flickr.com/people/argiolus/), on Flickr

Nate Potter, Austin TX

rdenney

11-Feb-2012, 22:44

I would not want the light source under anything, especially if it's a halogen source. I suspect that's why the diagram shows it to one side--heat management.

I still like the idea of a horizontal bed with a vertical film stage. That does complicate the film-stage movement, but not by that much.

I have just bought a couple of precision positioners and will report more when they arrive. These are used by the macro photography dudes, and Peter turned me on to them. I bought a pile-o-parts lot on ebay that I could tell from the picture had two of them in the pile. These lack the range of motion to serve as the x-y stage, and I bought them to use for precision focusing of the camera. But they can be obtained in longer lengths. New, they are expensive (hundreds), and I'll have to mount an Arca clamp to it, which I already have. I'll lock the two ends of the bellows to the same arca plate after I mount the lens and determine what length I want.

There are also precision sliders, but with these a locking mechanism will be needed. They are also expensive new, but they do turn up on ebay. They are used for just this sort of thing, but they might not have the precison of a microscope stage, though I don't think at 1:1 that level of precision is a requirement. My sense of these is that they eliminate lash greater than something like 0.0005", but that's speculation. I just bought a set that includes two sliders and about 20" of precision track--enough to make an x-y stage large enough for 4x5. I'll know more when it arrives. I'll need to devise a locking method, especially if the stage is vertical. But a vertical stage has a couple of advantages: 1.) the film can be vertical, which means it will stay flat without needing the sheet of glass. I intend to use a D2 negative carrier, modified and attached to the sliders as necessary. 2.) all the components can be mounted on a single massive flat surface, like the granite sink cutout I have sitting in the garage. 3.) no mirror is needed in the light source--rising heat will not be aimed at the film stage. I will probably try to use the condensers from my D2 enlarger, and maybe its light source, too. In fact, I might just mount the head sideways on the bed.

The baffling is pretty easy with black construction paper and adhesive flocking material, which is cheap at Edmund Scientific. The objective is to prevent any light that is laterally escaping the collimated path to get back to it.

So far, I've spent too much: $59 for the positioner, $95 for the two sliders and the 20" of rail, $15 for a 62mm filter stacker, $15 for a Pentacon Six reversing ring (to which that female 62mm filter stack cover will screw, to serve as a lens board on the Pentacon bellows, which I already had), and $75 for the LiNOS lens, which I still have to mount. So, with a light source already in hand and nothing built yet, I've got a quarter-kilobuck into it. New, I think I'm looking at no less than a coupla-grand of parts, so this is not a cheap proposition unless you can source the stuff on the used or surplus market. I don't think this is something that can be produced beyond a set of plans and some clues, but anyone making one of these will have to made do with what they can find.

Back to those condensers--this might be a good use for used small-format condenser enlargers that are often sold for shipping or given away. Probably cheaper than buying those condenser lenses from Edmund.

I'm still not sure the stitching will work. But if this is a bust for 4x5, I'll use it to scan a bunch of 35mm slides, which it will do much faster than my Nikon, and with probably good enough quality. So, I have a story to tell to the property manager if there is an audit.:eek:

Rick "whose auditor doesn't read this forum" Denney

Tobias Key

12-Feb-2012, 04:48

I was looking at this thread and wondered whether it might be possible to adapt a monorail camera as a scanning device? It would have a number of advantages, you'd have two parallel standards, gearing and shift and would be cheap to buy. You could remove the ground glass from the rear standard and hollow out a normal film holder to hold the film. Chinese manufacturers already make adapters to fit DSLR's to LF cameras, all that would be needed would be to create a lens mount on the other side. I'd suggest this as it could be easily marketed as a kit. A ready made film holder to match the dimensions of a double dark slide (newton glass would be possible) and the DSLR mount that could be supplied in common lens board sizes. This could be manufactured in small volumes profitably, and could feasibly sell for much less than a V700. Users could choose the number of frames they wanted to use in their scan, and their own light source.

Nathan Potter

12-Feb-2012, 08:16

Rick, indeed heat is an issue when using a vertical arrangement with the bulb at the base. Although I've used a small 12 volt halogen bulb driven by an adjustable microscope transformer (about 3 amps), so total power of around 36 watts. Even a brighter source would be one of these goose necked illuminators you see on _bay occasionally - also halogen and intensely bright where the end of the goose neck is essentially a psuedo point of say 1/2' diameter.

A horizontal arrangement has advantages that you point out especially for proof of concept in that it is in the form of the tried and true optical bench which provides precision in the alignment of the optical axis and maybe natural film planarity.

For the ultimate in image quality I think one would want to wet mount the film to glass anyhow. A vertical X/Y stage is no problem in the horizontal direction but needs to hold position in the vertical so a stop mechanism or extra tight races would be needed.

Ahh, condensers from an old condenser enlarger - had not thought about that. As long as they are defect free that might be ideal for a functioning prototype.

Tobias, you would need enough rise/fall and shift to cover 5 X 4 with a monorail approach. And I'd worry about the stability of the frames as well as acceptable planarity in the standards.
Then you still need a light source which I still think may need to be collimated to control scattering degradation. Nevertheless that could be rigged for another proof of concept experiment.

Nate Potter, Austin TX.

Peter J. De Smidt

12-Feb-2012, 10:07

These are great ideas. This project has the greatest chance of success if we get lots of good input.

Right now, I'm procrastinating about going out into the unheated shop to finish my first reasonably complete prototype.

Regarding a collimated light source, would the following work? Take a piece of aluminum flashing, such that it'll cover the front of a flash. Make a round hole in the flashing directly over the front the flash reflector such that light is only let out that hole. Would that be a possible replacement for a halogen light in Nathan's sketch?

Another option would be to use an LED. A high intensity 3w Cree would probably produce enough light.

Ben Syverson

12-Feb-2012, 10:34

One thing to keep in mind when it comes to light sources... Ideally the color temperature of the light will closely match the "native" WB of the sensor. For most sensors, I think this is close to daylight. As you move to other color temperatures, you increase noise in the color channels.

Another option for color negative is to gel the light until the orange base of the negative is photographing as a neutral gray. That way you're not having to subtract the mask digitally. The less you have to push the image around digitally (whether in-camera via WB or on the computer via Photoshop), the better.

Bob Salomon - HP Marketing

12-Feb-2012, 10:37

Peter J. De Smidt

12-Feb-2012, 12:00

Mike Anderson

12-Feb-2012, 12:28

Bob,

I appreciate the suggestions.

The Kaiser Prolite Scan Lightbox 17"x19" looks very nice, which it should be for $2000!
http://www.adorama.com/KRPSLB.html

The Gepe Pro 8" x 18" Slim Pro Illuminator Light Box with A/C Adapter, 1.5" Thick is more reasonably prices at $165, but I don't know how even it's illumination is. See: http://www.adorama.com/GPLB818.html

Seems like there's 2 ways to go with respect to evenness of illumination:

expensive hardware to provide even illumination
cheap hardware with uneven illumination and correct it in software

Peter J. De Smidt

12-Feb-2012, 13:17

I'm hoping for:

3. cheap hardware with less than 1% variation in light intensity.

I'm close now. It shouldn't be that hard to achieve.

Nathan Potter

12-Feb-2012, 14:11

Peter, a pinhole intercepting light would work but the intensity reduction would be extreme since most of the illumination flux is intercepted by the area outside the pinhole. The way to get around that is to use a double condenser system when using a broad area light source such as a non point incandesant or electronic flash. The setup would use a condenser closest to the flash to focus as much light as can be grabbed from the flash to a point (called a virtual point source) then recapture all that flux further on using a second condenser to correct the rays to infinity corrected (parallel rays as shown in the diagram). The idea is to capture and preserve as much of the original flux as possible through the optical train. While more complicated and difficult to work out experimentally from a uniformity angle using a flash appears to have a big advantage in the way of color fidelity (as Ben says) but also high speed exposure for vibration mitigation.

Hadn't thought about an LED source which could be interesting to investigate; although the spectrum of those show some high narrow band peaks which I believe are difficult to color balance well. In fact the tungsten source would need a CC filter to balance to the recommended color temperature for the particular DSLR sensor - but that should be straight forward.

BTW it would actually be informative to see results from both a diffusive light source and a collimated light source so I wouldn't want to discourage both approaches for a feasibility experiment. Geez, I wish I had some time to fool around with this.

Nate Potter, Austin TX.

Peter J. De Smidt

12-Feb-2012, 14:22

There's some Cree LED fixtures that are over 90 CRI. I'll look up the link later. Here it is: http://www.cree.com/products/modules_lmh2.asp
See also: http://search.digikey.com/us/en/products/LMR020-0650-40F9-10100TW/LMR020-0650-40F9-10100TW-ND/2524694

Sawing thick hard maple by hand is tiring. I need a break. Whew!

peter ramm

12-Feb-2012, 14:25

Peter, here are a few links you may find interesting. This stitching system is for microscopy, but discusses some general issues.

http://www.mcid.co.uk/applications/montaging.asp

This macro equipment link is just to show an illuminator that is stable (important because you will be acquiring images over time and don't want brightness variation) and even.

http://www.mcid.co.uk/hardware/map/

The illuminator was about $3000 new but they come up used. For example (no idea what they want for it):

http://www.usedlabequipment.com/lab_equipment/Northern+Light%20R95+Transilluminator+Light+Box_18225.php?DepId=&CatId=&SubCatId=&cPage=1&keywords=northern+light&ManId=

Moving the film is the problem. Ideally, you would just tape it in place on the illuminator and then move the whole thing under the camera on an XY translation table. Some precision is required because any angular errors will kill your correction software. There are gliding stages that you could mount an illuminator on, but expensive I think. eg

http://www.solino.com/cms/index.php?article_id=178&clang=0

Don't know if something like this would work.

http://www.visionaid.co.uk/product_view.php?pid=80011

Theoretically, as has been pointed out, a collimated light source is better but implementing one that doesn't actually make things worse takes some skill. Diffuse is easier. Generally, we match collimated sources to scanning optics and diffuse sources to area arrays. A mask can help the flare somewhat, especially if you pick a lens that focuses very close to the film. My favorite lens for this type of thing was the 55 MicroNikkor, which is pretty cheap used I think.

I am enjoying following your progress.

Peter J. De Smidt

12-Feb-2012, 15:02

Hi Peter,

Thanks for the links. I'll check them out.

I'll be using a 55mm Micro Nikkor for my first tests.

marfa boomboom tx

12-Feb-2012, 15:17

horizontal, not vitical... broad source, not point source, move the film, not the camera.
use automated movements and software to do the bulk of the work.

scanners use a cold cathode lamp.

I have an xy system running, and am confirming the accuracy ... how repeatable it is. The software is home rolled, based upon "tiling" or "mosaicing" =-- for those of you with interest in your own path.

I have my own implementation of SIFT (I've never seen their code), but since SIFT is patented, this may not be a "strong" path for software implementation.

my costs: 5 hrs software // free developer tools //
arduino based xy system with steppers ~$240 ..
the steppers can be $15 items since the load of moving film is low. I've gear boxes that are giving me finer gradation of movement. My goal is repeatability though.

film holder, on hand "dental clip" stainless film hanger

Oh, the arduino system can trigger my Nikon ... next step is a "shake" monitor....

Peter J. De Smidt

12-Feb-2012, 16:28

While flatbeds usually use cold cathode lamps, film scanners sometimes use other sources of illumination. For example, Nikon's Coolscan scanners use leds.

My experience with various scanners indicates that collimated sources, ala a Coolscan, work best with fine-grained negatives, whereas diffuse sources work best with coarser-grained negatives. For awhile, for example, a gentleman was selling diffusers, a ScanEnhance, for Milnolta's MutiScan. It'll be interesting to see if those results are the same with DIYS. One nice thing about this is that everything is modular. You could easily swap the sources for different types of film.

A gentleman just made a vibration sensor at the macro forum. See: http://www.photomacrography.net/forum/viewtopic.php?t=15941

I'll be very interested to talk to you when we get to the automation stage.

Mike Anderson

12-Feb-2012, 17:53

...move the film, not the camera..

Are you moving the light source with the film?

Peter J. De Smidt

12-Feb-2012, 17:55

The light source, camera, and lens remain stationary. Only the film carrier (and film) moves.

Tobias Key

13-Feb-2012, 05:18

A gentleman just made a vibration sensor at the macro forum. See: http://www.photomacrography.net/forum/viewtopic.php?t=15941

You can get vibration meter apps for android and iphone - I've downloaded one to see if it will help with setting up tripods in windy conditions or to judge the effects of shutter vibrations.

Works wells enough on a pentax 6x7 prism, the shutter vibration measures 4.5 on the richter scale!

rdenney

13-Feb-2012, 05:53

The precision slides I've bought, but have not yet received, are on page 1099 of the McMaster-Carr catalog, part no 8438K3. The slides use captured ball bearings for alignment. The track is fully machined and expensive, at $.51 per millimeter when bought new. The carriers are $75 new. These are designed to move stuff easily and precisely, and might be the source parts for the XY tables, which have the problem of being opaque. I bought two carriers and 500mm of track in a surplus bundle. Considering the light weight of the negative holder, I think it will be fine to cantilever the holder off one side of the carrier. I'll know more when it arrives.

As far as light sources go, it's easy enough to obtain a stabilized power supply. This is especially true if using an LED source, which can be powered by well-regulated DC power supply.

I'm familiar with the microscopy application--that's what I discussed with my computer imaging expert buddy, whose wife is an ophthalmic researcher. They have issues with what they are doing, but then they are looking at things at 100x. We are looking at stuff at 1x. Let's not overdesign this before we really know what we need.

By the way, a good source of precision straightedges and squares that can be used as parts, visit your local Woodcraft.

My engineering design sensibilities suggest that a vertical arrangement perches the weight of the camera on the end of a cantilevered column. Even if you brace the column (as I always did my Omega enlarger), the camera itself and its focusing adjusters will be cantilevered out over the target. If I'm going to cantilever something, I'd rather cantilever that lightweight film stage, which can be made quite stiff even with very lightweight trusswork or stiffeners. I'd prefer the tracks carrying heavy stuff be loaded normally (normal meaning at right angles to the axis of movement of the track), rather than applying a bending moment on the track. That means the tracks for the heavy items need to be horizontal. If I'm going to load a track in bending, I'd want that load to be the lightest load in the arrangement. That also allows me to fasten the camera directly down to my heavy base.

And why would I need a fluid-mounted negative? Just to minimize dust? Funny, but I never needed that in an enlarger, even making substantial enlargements. I always used a glassless carrier for 4x5, and it was horizontal which could allow the film to sag. I never had a problem with losing grain sharpness even into the corners, and I was using an old 139mm B&L tessar Ic barrel lens (circa 1948-50) as an enlarging lens.

But that means the vertical-movement part of the negative stage will need to be on a track that can be locked in place easily and without upsetting it, and I have not solved that issue yet.

There may be an advantage to a little contrast reduction, by the way. It might reduce an aliasing problem. But I'm not trying to match a drum scanner at this point. My quality target is a high-end flatbed like a Creo/Scitex.

Rick "visualizing designs" Denney

Peter J. De Smidt

13-Feb-2012, 06:39

marfa boomboom tx

13-Feb-2012, 07:13

remember, this can't be done by small scale folks :) nor can you make an electron microscope in your garage.. http://www.youtube.com/watch?v=VdjYVF4a6iU&feature=player_embedded

rdenney

13-Feb-2012, 08:42

For a vertical negative stage, you could use a piece of flat magnetic metal with a hole cut for the light source. See: http://www.kjmagnetics.com/ for a nice choice of magnets. For a negative carrier, it could be a clamshell design, with either a solid and clear AN back glass, or glassless. The holder could use magnets to hold it to the flat piece of metal, and those could also probably hold the holder shut.

I'm still going vertical, at least for now. My Velmex slide is rated for 50 lbs with a vertical load, and my enlarger works ok with a cantilevered load.

I dug down into the manufacturer spec-sheets for the slides I bought, and they are IKO C-cube Linear Way slides. Each slide has a series of retained ball bearings that run in a machined track on the rail. It's all very tiny--the surface of each slider is about an inch by a little over an inch, and the track is 9mm wide. But it is set up for zero clearance on the balls, and the rated accuracy is 4 microns over 125mm of motion. The maximum torsional load to maintain that tolerance is 5.6 foot-pounds. The load will be an Omega 4x5 film holder (glassless), which may weight several ounces. I think I'll be okay.

I'll mount one slide flat on the surface, and mount a right-angle structure on that. I'll bet I can find a carpenter's jig of some sort that will serve admirable. I'll mount the vertical slide on that. The rail is carbon steel, so I can use the magnet as a movable stop for vertical motion.

The Omega carrier is aluminum. But it holds the negative on all four edges, and includes alignment pins to keep it in position. I think I'll use strong clippies to hold the two halves together. Only one half needs to be fastened to the slide.

I'll mount that first, and then use a square to mount the Velmex focusing track for the camera. That should ensure plano-parallism.

Fastening it all to the granite is no mean feat. I might start with something like an oak cutting board--a thick one. If it all works, then I can fasten that down to the granite for mass stability, and put all that on Sorbethane bumpers if I want to brag about stability.

Really, I'm least worried about the light source, and most worried about stitching.

Rick "getting sucked in by the machine design challenge" Denney

Peter J. De Smidt

13-Feb-2012, 09:03

remember, this can't be done by small scale folks :) nor can you make an electron microscope in your garage.. http://www.youtube.com/watch?v=VdjYVF4a6iU&feature=player_embedded

:eek:

Really, I'm least worried about the light source, and most worried about stitching.

Yes, me too.

I'm really looking forward to seeing your system, especially the negative stage.

I just finished building my negative stage, and now it's on to putting the camera support stage together. But first, Lily is insisting that we go to the dog park.

Bob Salomon - HP Marketing

13-Feb-2012, 11:12

Bob,

I appreciate the suggestions.

The Kaiser Prolite Scan Lightbox 17"x19" looks very nice, which it should be for $2000!
http://www.adorama.com/KRPSLB.html

The Gepe Pro 8" x 18" Slim Pro Illuminator Light Box with A/C Adapter, 1.5" Thick is more reasonably prices at $165, but I don't know how even it's illumination is. See: http://www.adorama.com/GPLB818.html

The Gepe box is more then even enough for medical xrays which is our biggest market for them. Depending on the size, for dental (up to full mouth pans), vetenary and MD work.

Nathan Potter

13-Feb-2012, 11:32

Peter, now is not the time to go to the dogs! Sounds like you folks are having fun with this interesting project.

Nate Potter, Austin TX.

BetterSense

13-Feb-2012, 12:46

This sounds like a fun project, which I think I could do except I don't know anything about stitching software. Does anyone know any good, scriptable tools that can do switching? Can you do it with ImageMagick yet?

buggz

13-Feb-2012, 14:10

I have been using Microsoft ICE with very good results.
AND, it's free!

This sounds like a fun project, which I think I could do except I don't know anything about stitching software. Does anyone know any good, scriptable tools that can do switching? Can you do it with ImageMagick yet?

Cesar Barreto

13-Feb-2012, 15:20

"Really, I'm least worried about the light source, and most worried about stitching".

A couple of years ago I made a series of "scanning" using a Nikon camera, up to 36 captures for 5x7' negatives and really had some trouble about stitching.
When one starts making small bits from a big negative, often it gets to featureless areas where stitching software may have a hard time joining each part.
I just used PS5, wich works wonders on areas full of details and texture, but may fail miserably when dealing with skies and alike.
A poor man solution was writing some small nankin signs on the back of the film, so software could find something useful for it's recognition tools. Simple retouching got rid of them in the final images, while a wet cotton swab cleaned the films.

Cesar Barreto

13-Feb-2012, 15:44

Last month I bought a Multiphot camera and I think it could be a great tool for this job.

For those who never heard about this fantastic Nikon: http://www.microscopyu.com/museum/multiphot.html

Peter J. De Smidt

13-Feb-2012, 18:47

Yeah, those are pretty neat. The lenses are supposed to be very good.

Nathan Potter

13-Feb-2012, 19:29

Last month I bought a Multiphot camera and I think it could be a great tool for this job.

For those who never heard about this fantastic Nikon: http://www.microscopyu.com/museum/multiphot.html

Cesar I used to use that very model for industrial work - very stabile. I have the equivalent made by Leitz for 4X5 format with the most amazing 16 bulb ring light and power supply. It is very sturdy, essentially a vertical optical bench with a precision sliding optical rail. I've adapted this unit for macro work using 35 mm film cameras in the past. It would be ideal for the task at hand here for an experimental setup. The mint Contax IIa just unboxed from the original packaging is not part of the Leitz system. Picture of the Leitz version below.

Nate Potter, Austin TX.

http://farm8.staticflickr.com/7065/6873397171_01d395ba1f_z.jpg (http://www.flickr.com/photos/argiolus/6873397171/)
LEITZmacro (http://www.flickr.com/photos/argiolus/6873397171/) by hypolimnas (http://www.flickr.com/people/argiolus/), on Flickr

Daniel Moore

14-Feb-2012, 00:36

Would applying a sheet of polarizing material over a light source improve it's quality for this purpose?
Tried to read the whole thread tonight and couldn't so forgive me if I missed this.

Also, realizing there are going to be cheaper solutions, I did want to at least mention this little guy at 295.00 (http://www.edmundoptics.com/products/displayproduct.cfm?productid=3358). There are others at Edmund Scientific but it seemed best suited.

Cesar Barreto

14-Feb-2012, 01:50

peter ramm

14-Feb-2012, 07:19

Right now there's a Multiphot camera offered on the bay for $6,5K, but I had the chance to buy one for a fraction of this, down here in Brazil.
I haven't tested it yet, but it needs a new bellows and I need to adapt a power supply for their 6w bulbs.

I understand there are some stages used on microscope work wich moves the samples on X and Y axis, but I never worked with those things and I guess someone might know if they could be adapted as scanning devices.

Cesar, microscope stages are made for small movements. The largest I am aware of handle microwell plates (about 10 x 7 cm). You might find one that handles 4 x 5" scanning, but the retail on something like that would be upwards of $10K. Never mind the software that runs it.

There have been various stand-mounted macro photo rigs, all very pricey and not really suited to film scans (the illumination path is not optimal). All the dedicated macro systems I know of predate desktop imaging (no stitching correction) and I don't know how well modern ones solve stitching. I have seen modern stereo microscopes used in this way but can't remember how well the software did. Really, macro systems are designed to image single fields at low power, not stitch at higher power.

It would be fun to build something like this starting with an optical bench, scanning mirror system, laser and a rail-mounted PMT package. However, I think the goal here is to make something about the same cost as an Epson, so the fellows are limited in their parts selection.

rdenney

14-Feb-2012, 08:19

The thread has grown so long people are unlikely to read all of it. So, while we gather parts and build prototypes, which will lead to posts that can provide a summary of what was done, it might be useful to reiterate the requirements.

Needs:

The user will scan 4x5 or (possibly) 8x10 negatives, using the DSLR they already own, and additional equipment that can be sourced surplus or used for less than the cost of an Epson V750. The user will use the scan to make prints up to 8x enlargement (it's just a number larger than 4--which the Epson can do). The user will make use of the apparatus for at least 10 years.

The user is not a software programmer and will use products generally available in the market or openly provided from a source that will maintain it.

Some users may add this: The user will start the scanning process and return when the final images is scanned and ready for editing. (I don't add this, because I can live without it and I think it leads to requirements I'm unable to fulfill within my budget, or software requirements that are incompatible with other requirements, particularly longevity, except for those who are software programmers.)

Requirements:

The system shall provide an MTF of at least 50% at 2880 spi (based on 8x enlargement printed at 360 ppi).

The system shall use a standard DSLR, with a standard macro lens arrangement.
--The DSLR shall provide a minimum sensel spacing of 8.8 microns.
--The macro arrangement shall provide a 1:1 magnification.
--The lens shall provide 50% MTF at 60 lines/mm over the entire frame.
--The lens shall provide distortion less than 0.04% (less than one sensel across the frame)

The system shall provide a camera fixing and focusing apparatus rigid enough to prevent more than 4.4 microns movement during the exposure.

The system shall provide a focusing adequate to fulfill other requirements.
--The focusing system shall not alter the magnification

The system shall provide a movable negative stage.
--The negative stage shall allow sufficient movement to scan the entire 4x5 (or 8x10) film
--The negative stage shall hold the film flat enough to fulfill other requirements
--The negative stage shall provide movement precision within 5 microns over the range of movement
--The negative stage shall provide manual movement OR The negative stage shall be moved by an automated system (which will entail other requirements I'll let someone else write)
--The negative stage shall be parallel to the sensor plane within 0.04%

The system shall provide a light source.
--The light source (or correction system) shall provide even illumination within 1% across the frame
--The light source (or correction system) shall provide even illumination within 1% from frame to frame.
--The light source shall support MTF requirements.

I'm sure I've missed a few.

In reviewing these, it seems to me that negative stage has to be precise indeed to avoid imposing geometric distortion that would complicate the stitching process. And the specified lens may be unobtainable for a reasonable price. But if we don't write requirements, we won't fulfill them.

Based on these, I've bought another pair of the precision slides, because I think I'm going to need two slides each for x and y to minimize any cantilevered flexibility in order to fulfill this requirement. That is going to put a strain on my construction skills, to be sure. And I'm now way over budget. But a design is forming in my mind.

Rick "practicing his systems engineering beliefs" Denney

William McEwen

14-Feb-2012, 08:54

The thread has grown so long people are unlikely to read all of it.
Rick "practicing his systems engineering beliefs" Denney

That's me.

Forgive me if this has been mentioned here...

I saw a video of Douglas Kirkland, who shoots some 8x10.

He places his negative on a light table and shoots a picture of it with his DSLR.

Seems to me we should invest in light tables and copystands, make hybrids, and go that route...

rdenney

14-Feb-2012, 09:06

That's me.

Forgive me if this has been mentioned here...

I saw a video of Douglas Kirkland, who shoots some 8x10.

He places his negative on a light table and shoots a picture of it with his DSLR.

Seems to me we should invest in light tables and copystands, make hybrids, and go that route...

So, do you think that would fulfill those requirements? The point is to provide a durable replacement for high-end flatbeds (at least) into the future. It's probably gonna take more than a plain light table and a copy stand, though I have no doubt that's fine for Internet display. Then again, the fixtures we are contemplating are just copy stands and light tables, with two exceptions: Adequate precision to fulfill the requirements, and the ability to move the negative precisely to allow stitching multiple images.

Rick "preparing for a film-scannerless future" Denney

Peter J. De Smidt

14-Feb-2012, 09:14

In general, I agree with Rick's summary, although I'm taking his word for it on some of the specific requirements.

One where I'm not so sure about, though, is:

"--The negative stage shall provide movement precision within 5 microns over the range of movement"

If you mean x-y positioning of the negative, we might not need to be so precise.
A DX sensor measures 23.6 x 15.8 mm. An overlap of about 25% should be enough for stitching. This would mean moving the negative 11.85mm moving across the y-axis for the second exposure. A 4x5 negative is roughly 95.25mm wide, which will take 9 frames to cover the width of the negative. 9 times 11.85mm equals 106.65mm. A total distance of 106.65 mm minus 95.25 equals 11.4 mm. So the total slop in the system needs to be less than 11.4mm for the width. Doesn't this mean that + or - 1mm per step would be ok, at least for 4x5" negatives?

That's with an automated movement. With manual movement to reference marks, wouldn't + or - approximately 3mm to each reference mark be enough? Take exposure. Slide carrier to mark. Make exposure...slide...

rdenney

14-Feb-2012, 09:27

If you mean x-y positioning of the negative, we might not need to be so precise.

Yes, you are right. I really mean in terms of distance to the camera, but also in terms of rotation to avoid the stitching having to rotate the pixels (which is lossy). But lateral position doesn't need to be precise, if we use software to find the stitch line, as I hope we can.

In the design I was visualizing, my negative holder would be like a flag mounted on one end to the slider. Precision is good in all axes except what I can only describe as the flag rotating around the flagpole. Doing so would cause perspective distortion, not mention loss of focus. Writing the requirement in that way, and being clear about what sort of precision might be needed for planarity to avoid distortion, forced me to put another slider on the other end--the flag is too big and catches too much wind to be that securely supported from one end.

I sure hope the stitching works because I've spent too much on the machinery already.

Rick "probably chasing a pig in a poke" Denney

marfa boomboom tx

14-Feb-2012, 10:43

Some thoughts on the Software:

My best guess for long term users not interested in rolling their own,
even from source code (whatever that is:)-- 1,2,3 have source and active community of coders.

1- Hugin
http://hugin.sourceforge.net/tutorials/index.shtml

2- Or panotools for a panoramic approach. (the stitcher isn't sourced,so ?risk??)
Panoramic approach
http://wiki.panotools.org/

3- For those who like command lines and think a Stitch approach can be solved in your situation, this is worth your time:
http://www.imagemagick.org/script/index.php
see: 'montage' examples

4- Finally, there is the GigaPan thing. It costs, but gigatools may be "free", I don't know. The gigamicroscope stuff depends upon the software from GP.
Of course, that whole thing is only as stable as the University System and Carnegie Mellon is. Though, it seems the gigapixel imaging phase is picking up advocates, not losing them.

//

mortensen

14-Feb-2012, 10:50

Thanks for the resume, Rick - a good sum up for those of us who didn't have patience to read 42 (!) pages in here.

Now, post some blueprints and prototype pictures :D

Nathan Potter

14-Feb-2012, 12:20

X/Y stages have the advantage of precision in planarity over a large area (and of course also precision in X/Y positioning accuracy). It has been suggested that the X/Y positioning accuracy is not needed as long as rotational errors are not introduced during translation and I think this is true depending on the completeness of the stitch algorithms. So the planarity in the Z direction is what we want and should not want to pay for precision in X/Y unless that is unavoidable - and it may be.

In fact microscope stages come now in sizes up to 14 X 14 inches or more for studying IC wafers of 12 inch diameter. But we couldn't afford to pay for those and generally they don't come with a very large aperture for scanning over a light source. Dennys' approach of just buying the slides drastically reduces the cost but requires mounting a transmissive plate and a mechanical contrivance to paired slides (paired to maintain planarity and magnification factor as he explains) while scanning. The slides seem like a good experimental approach but need mechanical design and fabrication elements.

Ricks' Requirements write up is a quite proper way to establish performance objectives, in this case chosen to exceed the performance of an Epson V750 by a substantial margin. Certainly I would choose to achieve a nominal 10 um resolution at say 50% contrast but would be overjoyed to see 5 um resolution at 50% contrast as might be obtained using a new Nikon D800E just for an experiment. Budget requirements would keep the DSLR to a minimum of cost based on Ricks requirements.

Nate Potter, Austin TX.

rdenney

14-Feb-2012, 12:59

In fact microscope stages come now in sizes up to 14 X 14 inches or more for studying IC wafers of 12 inch diameter. But we couldn't afford to pay for those and generally they don't come with a very large aperture for scanning over a light source. Dennys' approach of just buying the slides drastically reduces the cost but requires mounting a transmissive plate and a mechanical contrivance to paired slides (paired to maintain planarity and magnification factor as he explains) while scanning. The slides seem like a good experimental approach but need mechanical design and fabrication elements.

Yes, that will be the tricky bit. But I have some ideas. It involves buying some precision carpenter's squares.

The problem I had with X-Y tables was that 1.) I couldn't find one with an adequate range of motion in the surplus market (read: ebay plus only a couple other places) and new ones were prohibitively expensive, and 2.) none that I saw had an aperture for allowing the negative to be backlit. The only solution to that would be to put the light source on the table, which makes the part being moved heavy again.

I don't think using individual slides will be that difficult with the approach I have in mind. But we'll see. The question will be their general availability to those who want to follow along--they are expensive new. But I had no idea what was out there last week, and am designing based on what I find. That may well be the price of entry into this game, until someone can figure a way to productize it.

Rick "making it up as he goes" Denney

marfa boomboom tx

14-Feb-2012, 14:08

I used a couple of DISCARDED CDRoms to form my XY table.

see this for an idea; http://blog.whattomake.co.kr/158

and CDRs go for 1-3 dollars on ebay

Cesar Barreto

14-Feb-2012, 14:11

Rick did actually resumed pretty well the whole scene, but I would like to remember to those who already have Sinar P or equivalent 8x10 cameras and don't mind to make the scanning with hands, instead of machines, that X and Y movements works very fine with these cameras and both planes are naturally aligned.

A mask for holding 4x5' or 5x7' negatives set on the first frame and a nice SLR digital camera installed in the back and that's it. Of course some variations are possible and I made my scans using a Kaiser column for up and down movement and the Sinar for lateral shift, since at this time I just had a 4x5' camera. Aligning was a PITA, as you say, but after some trials it was easy to go along and make a bunch of scans each day.

Peter J. De Smidt

14-Feb-2012, 16:25

I used a couple of DISCARDED CDRoms to form my XY table.

see this for an idea; http://blog.whattomake.co.kr/158

and CDRs go for 1-3 dollars on ebay

Nice idea! I probably have 4 old defunct cd players sitting around.

AFSmithphoto

14-Feb-2012, 20:11

I was just given a Nikon SB-28 Speedlight. It goes down to 1/64th power. I'm considering making a light source out of white styrofoam and difused plexi for this experiment. Does anyone know how repeatable the intensity of these flashes are? I guess I'll find out. Maybe I'll trying using one of the mixing boxes from my De Vere 504 for the light box.

Just started reading through this thread, so for all I know my input is coming to late, but I would recommend poping the flash a couple of times before you take the first image.

As a lighting professional, it is my experience that anything other than an old fashioned continuous filament needs to "warm up" before reaching consistent output. In this case, the warm is literal. Everytime you fire the flash it produces a small amount of heat, which is unlikely to effect exposure much, but could theoretically effect color temprature just slightly enough to blow your results. Good luck.

Ben Syverson

14-Feb-2012, 21:16

I'm trying to hold my tongue further until more tests are performed, but my strong suspicion is that strobes will not fare well for this application. The ones I've used vary by about 1/20th of a stop from pop to pop, which is virtually nothing but can put a real damper on animation or stitching.

It's telling that the stop motion animation films shot on DSLRs in recent years all used continuous light. The camera's shutter is much more consistent.

Daniel Moore

14-Feb-2012, 22:48

Most stitching software I've used recognized slight exposure variations and compensated for them in the process of blending, otherwise none of the stitched images with sky areas would be worth a damn. I wouldn't discard flash illumination just yet.

Tobias Key

15-Feb-2012, 02:39

Just as another aside but does anyone have any experience or knowledge of using a Bowens illumitran slide copier? I know they were designed for making a single 35mm slide from an original but the lightbox design might be useful as a reference, or even an old illumitran could be used as the basis for a design. Note the illumitran did use flash for the exposure and contained a contrast control unit. Anyway I thought it might be worth looking at to see if there's any ideas that could be used.

There is one pictured with a 4x5" lightbox so if there was a way to incorporate a sliding mechanism just above that you'd be set as long as you could adapt your DSLR to fit.

Also if anyone had a lot of 120 or 35mm to be digitised this looks like an interesting solution

http://members.bitstream.net/~tlmartin/copiers.html

http://members.bitstream.net/~tlmartin/dslrfaq.html

rdenney

15-Feb-2012, 06:21

Just as another aside but does anyone have any experience or knowledge of using a Bowens illumitran slide copier?

I shopped for one recently in the hopes of snagging that bit of obsolete technology cheap. But the word seems to be out for people wanting to digitally archive slide libraries. The last few I've seen sold for many hundreds of dollars.

Rick "hoping to find one in a garage sale someday" Denney

Peter J. De Smidt

15-Feb-2012, 06:44

Yeah, Illumitrans are going for quite a bit currently. I used to have a different brand, but I can't remember which one.

Tobias Key

15-Feb-2012, 06:52

I shopped for one recently (or a Honeywell Repronar) in the hopes of snagging that bit of obsolete technology cheap. But the word seems to be out for people wanting to digitally archive slide libraries. The last few I've seen sold for many hundreds of dollars.

Rick "hoping to find one in a garage sale someday" Denney

Maybe someone should tell Bowens and they'll put them back into production!

sethlatimer

15-Feb-2012, 08:26

This is a great thread and I am really interested in where it is going. I am beginning to get worried about the structure organization of using a discussion thread to facilitate this.

There have been helpful summaries of progress which help, but are a bit difficult to find. Also, we seem to rehash issues that have been addressed or are currently pending resolution.

Would everybody be better served if there were a wiki page (or more than one page) to document this progress? Free pages can be set up at wiki.com or other locations. For those who don't know a wiki is just a collaboratively edited document. We can document progress, establish next steps and generally work together a bit better. I have seen them used to pretty good effect in software design projects.

An important prerequisite is that the group would have to have almost unanimous buy-in on the idea and it would require everybody to check-in at one additional place on the interwebs. There are probably other issues I am not thinking of.

Let me know what you all think?
Seth

marfa boomboom tx

15-Feb-2012, 08:30

Light Source (LS1):

a--- (2) PH/140 bulbs + sockets and wiring
they are placed at opposite ends of the "reflector tunnel [b]" ~$15

b--- (1) length of scrap conduit 4"x~ 7" less than One Corona
grind out a 1/2 inch slit along about 6" of one side -- the remaining Corona.
[ I think this is called sched 40 ]

c-- (1) krylon mirror spray $11

d-- diffusion cloth... one used white hanky.
will experiment with other clothing items after checking out
the local university laundry day.

LS model II:

I put my dichroic color head on its side and used it.

Note: Obviously I headed the diffuse route. One "strip" light, rather than a point light.
My source target was first a 4x5 color neg. I changed over (to protect it, initially) to a toss
away Kodalith negative, complete with "pin holes." .... this was a lucky accident, in that those holes provided me very fine targets
to verify my stepper controls. It also gave me the idea of "pin hole" positioners along the XY paths.

Pinholes and scratches, the bane of photographers everywhere, certainly provided me some luck along this path.

rdenney

15-Feb-2012, 08:45

This is a great thread and I am really interested in where it is going. I am beginning to get worried about the structure organization of using a discussion thread to facilitate this....

An important prerequisite is that the group would have to have almost unanimous buy-in on the idea and it would require everybody to check-in at one additional place on the interwebs. There are probably other issues I am not thinking of.
I can't think of how that would help people build prototypes at this point. I do think might be valuable for those who are actually performing specific tests, who have specific contributory expertise, or who are building prototypes, to have a separate space where they can discuss details without having to rehash things for people who are unwilling to track through the long thread.

A blog might be better, where blog entries track progress and the commenting peanut gallery is confined to the comments.

But the thread has become place to discuss details for those who have participated in the thread all along. When we actually have results to report, it might be better to take a few items out of this thread and create a new thread that provides a summary. I suspect someone will want to document that on a web page, too, to open it up.

I don't think consensus is that important--each builder is starting with a variation on the requirements and going their own way. We really want a library of different approaches rather than one collaborative single approach. These will necessarily be home-built and each person brings a different set of objectives, skills and raw materials. I think we'd fight over Wiki edits.

Rick "not ready for configuration management" Denney

Peter J. De Smidt

15-Feb-2012, 08:58

Eventually, we'll do something along the lines that Seth has outlined. Perhaps we'll put an article on Tuan's LF page, or we could have a DIY forum, with an official build thread stuck to the top. Perhaps there could be a thread for each element, light source, lens, x-y system.... But for now I'm going to concentrate on getting the prototype done. I just made a mistake notching wood, and so it's off to the store, darn it.

Daniel Moore

15-Feb-2012, 22:27

Perhaps helpful in prototyping: http://www.emachineshop.com/
I haven't seen anything to suggest their software can import from a common preferred CAD program (ViaCAD!) yet.

Daniel Moore

16-Feb-2012, 11:03

emachineshop.com software can import .dxf drawings.

Wish I had more to offer this project.

Peter J. De Smidt

16-Feb-2012, 11:14

The first fully-functioning manual prototype is done. I'm going to take a bit of a break and then install and align it. I'll try to get some pictures up later today, although I'm not sure I'll have time to mess with stitching.

sraichur

16-Feb-2012, 11:59

Awesome peter,

Can't wait to see the results!!!

rdenney

16-Feb-2012, 12:14

Peter J. De Smidt

16-Feb-2012, 12:52

http://i955.photobucket.com/albums/ae37/peterdesmidt/side_view_2.jpg

sraichur

16-Feb-2012, 13:16

Peter,

Great job!

When can we see some initial '1 photo' scans, compared to the drum scanner?

Thanks

Peter J. De Smidt

16-Feb-2012, 14:32

I don't have a drum scanner.

Nathan Potter

16-Feb-2012, 14:54

Question: On the topic of collimated light sources, what about using a slide projector? The projection lens should certainly provide good collimation, and color correction filters could be mounted in 2x2 slides. I have never noticed anything resembling uneven illumination when lighting a screen with no slide in the slot. I even know where my projector is hiding at the moment.

Rick "curious" Denney

Hey Rick, hadn't thought about that. The intensity is there and maybe the uniformity is pretty good. I will say that it is still a somewhat extended source owing to the diameter of the illumination source (a broad area reflector). So the dramatic reduction in scattered light that can be achieved using a real point will not be realized. OTOH the nice tone modulation found in slightly more diffuse sources can be very appealing. It'll clearly be useful to characterize the various types of light sources because I think there will be preferences depending on ones' replication objectives.

Nate Potter, Austin TX.

sraichur

16-Feb-2012, 14:58

Peter

sorry, I though the 'Screen Cezanne' you mentioned in your blog, was yours?

Jim Michael

16-Feb-2012, 15:11

Hey Peter, that's an interesting looking negative carrier. How does that work? Nice craftsmanship on your rig.

Peter J. De Smidt

16-Feb-2012, 15:40

Peter

sorry, I though the 'Screen Cezanne' you mentioned in your blog, was yours?

No problem. Yes, I have a Screen Cezanne, but it is a professional pre-press flatbed. It has a ccd sensor, unlike a PMT sensor in a drum scanner.

sraichur

16-Feb-2012, 15:45

Oh ok, I thought it was a drum scanner.

Anyway, your prototype looks great! can't wait to see some scans.

Thanks

jb7

16-Feb-2012, 15:48

Very good Peter, looks nicely joined and sturdy-
Nicely finished too.

Must get around to photographing mine, and tuning it. It needs more work, mostly in evening out the light source. It's about three weeks old now, but I've been busy on other things...

Peter J. De Smidt

16-Feb-2012, 15:55

My prototype is made out of poplar and hard maple, mainly because that is what I had on hand or could easily get. All of the camera support framing is 1.75" (actual dimension) square poplar lumber, although the two feet are made of 2" square lumber. All of the joints are notched and bolted.

The Velmex 4000 series linear slide is stainless steel, and it's bolted to a slab of hard maple. (The maple was the cutout for for the glass in the negative stage.) I machined a mounting block out of Fortal aluminum to go between the Velmex and the Kirk Arca-style clamp.

If you have a live view camera, something like the Velmex slide would probably not be needed. I don't have live view, though, and so I went with the slide. I'm going to make custom extension tubes for each of the lenses I want to try, as I decided not to use a macro bellows.

The negative stage starts with a slab of laminated hard maple. It was the cutout for a sink on a commercial shoot, and the construction guys were nice enough to let me have it. I cut out an opening on the maple slab for a 12" square, 1/2" thick piece of plate glass. This is the surface on which the negative carrier will slide. Currently I'm using a negative carrier from my Cezanne, but if this works out, I'll make some custom carriers. The Screen holder is a clam shell, with AN acrylic on both interior surfaces. If I make a carrier, it'll be anti-Newton on the bottom but glassless on top.

There are two steel reference pins for placing an ABS sheet mask on top of the glass. The ABS sheet is cut such that the whole negative, in this case a 6x7cm negative, will be covered by sliding the carrier in the opening of the ABS sheet. There's a sliding arm to help locate the carrier when a side of the carrier is not touching the edge of the cutout in the ABS sheet. I'll make a specific ABS sheet for each film format that I'll scan.

The feet of the negative stage are threaded steel pipe, about 1.25" in diameter. The negative stage weighs about 50 lbs. The camera support structure weighs more. They could all be bolted or clamped to a support table, but I'm only going to do that once everything else is worked out.

Peter J. De Smidt

18-Feb-2012, 15:34

Stitching hasn't been too successful, so far, but I'm not an expert in it by any means.

Peter J. De Smidt

18-Feb-2012, 18:14

Daniel Moore did a stitch with PTgui for me. Thanks! It was 17 frames covering a 6x7cm negative.

http://i955.photobucket.com/albums/ae37/peterdesmidt/First_Panorama_web.jpg

The resulting file will make just under a 16x20" print at 300 dpi.

Apparently my calculations were a bit off. It looks like only 6 frames are needed with my Nikon 55m micro at an indicated 1:1.

jon.oman

19-Feb-2012, 10:40

To me, this looks good Peter! A bit low on contrast? From this image, I don't see any artifacts. Are you going to try a larger negative?

Ben Syverson

19-Feb-2012, 12:44

Looks great! Low-contrast is good, that's what we want (more information). You can always tweak contrast later.

Peter J. De Smidt

19-Feb-2012, 15:06

The image presented isn't very good, and it certainly isn't edited extensively. I thought it would be good, though, to see how the stitching handled the lines and concrete surface. With Daniel's help, it doesn't look like this is a problem.

There are a bunch of tweaks that can be made. In the first place, I'd like a slightly bigger light source. More importantly, though, it'll probably be very important to mask the lens very well, as contrast will suffer with extraneous light. I'll also build a negative carrier that has a cutout on top of the negative. Once those things are done, I'll start comparing the lenses I have. It looks like that for 35mm and medium format 2:1 might be better than 1:1, but only time will tell. Eventually, I'll get to trying bigger negatives, but at the moment that only leads to more work with little gain. Comparison with other scanners wouldn't make much sense until the system is fully tweaked out.

rdenney

19-Feb-2012, 18:56

Miracles sometimes occur, and I found myself this weekend with some time on my hands.

So, I've made much progress on my prototype.

The first picture (sorry for the iPhone pics of limited quality) shows the general arrangement--horizontal with the camera on a precision slide and the negative held vertically.

I have built two camera mountings (the second and third pictures), because the parts were available and each will provide a different feature. One is built with a Manfrotto hex-clamp, and the other with a Wimberley Arca-style clamp.

Both the slides came in a lot of laboratory equipment that I bought. The black one (second picture) is a typical adjustable dovetail slide. The nuts on the side are jam nuts for the adjusters, so this can be made to be tight with zero play. The slide had a riser machined into it, which provides a bit more elevated camera position. With the hex-plate clamp, I can use this one for cameras that don't have Arca plates or for when I'm lashing something up and don't have the right plate.

The camera is clamped to this stand right now--I am still awaiting a long enough Arca slide to allow me to lock down both standards of the bellows to the Arca slide. That will greatly improve the alignment and rigidity of the bellows. The bellows is made for the Pentacon Six, and I have a short extension tube plus a P6-EF adapter between it and the Canon 5D. The lens is an EL-Nikkor 105/5.6 mounted in a body cap and clamped into the front of the bellows. I will be improving that arrangement, but I'm still awaiting parts. The bellows are adjusted for 1:1 with that lens in this picture.

The bright aluminum slide is a Velmex 2500--very high quality. It's shown in the third picture. This one also provides a zero clearance in addition to no backlash on the focuser. I constructed an adapter to mount the Wimberley clamp using a 3/4" square tube. With this arrangement, everything is held with at least two fasteners, and there are no rubber pads--all surfaces are machined flat metal surfaces.

I mounted both slides on aluminum spacers to get the adjusters above the deck, and then mounted a couple of spare Sinar boards to those spacers. No actual Sinar board was harmed in this activity--these are cheapie Chinese knockoffs that I have since replaced with Sinar or Horseman-branded boards. I went to some care to make sure the boards are axially aligned with the slides, and I believe they are within 0.0005" over their length. (I have a machinist's square which has a rated accuracy of 0.000063", but I'm not exactly sure I believe it. I know I can't test it myself to that degree of precision.)

When I mount these on the horizontal bed, I will allow the Sinarish boards to slide between two aluminum angles used as a track. I will also mount an angle on the edge of the Sinar board to kiss those tracks. That way, I can slide the whole thing fore and aft to provide gross focus and then clamp the two angles together to fix the position.

The camera mounts really benefit from the horizontal arrangement I'm using. All the forces are applied across flat interfaces--very little is in bending. But that made the negative stage more difficult.

The negative stage uses an Omega D2 4x5 negative holder, which is two aluminum plates that sandwich the negative. One of the plates is attached to two IKO linear way slides, each of which runs on a vertical track on either side of the negative opening. These slides use a row of ball bearings that run in grooves in the track, and despite their small size, they are very precise. The track is machined hardened steel, and it's so hard that a hacksaw couldn't even scratch it. I had to cut the 20" section in half using an abrasive wheel on a Dremel tool. These slides have zero play. I fastened them to right-angle brackets that have machined surfaces used for clamp-ups for precision wood-working. They are machined to a right-angle accuracy of 0.002" over their length--but in practice they are a bit better than that (checked with that fancy machinist's square). Those two brackets are attached to slides on a crossing track for horizontal movement. Everything is true within a thousandth--but until I fasten down that side-to-side track, I won't know how stiff it all is. The holes are spaced on 20mm centers, and I'll put a screw in every hole.

The vertical negative will be sandwiched in the negative carrier, and being vertical, I hope there will be no tendency to sag. I am not using glass. No need to worry about fluid mountings, Newton rings, or extra layers of dust. Hopefully, I also won't need to worry about film flatness, but we shall see. Any Omega carrier can be clipped to this one, and the sandwich is held together with clips.

Note that the negative stage is at the bottom of its travel in the first picture, and at the top in the last picture.

All of this will be fastened down to a piece of phenolic-coated plywood that is made for tool decks, and that will be clamped to a slab of granite from a sink cutout. I will attached measuring strips to all the movements, and use spring clips to hold the negative stage in place at each position. That's the next step.

Rick "relieved to see some good stitches from Peter" Denney

Peter J. De Smidt

19-Feb-2012, 20:07

Looking good, Rick!

Darin Boville

19-Feb-2012, 20:21

Daniel Moore

19-Feb-2012, 20:50

A D5100 would make an excellent and modestly priced capture device.

Peter J. De Smidt

20-Feb-2012, 05:21

If we are looking at doing the scanning in batches, and not for a dedicated unit, why isn't this the way to go? If it is, what is the cheapest camera I could buy that would work? I would love to be able to scan 4x5's in this way, if better quality than a Epson 700.

--Darin

The challenges with a view camera are:
1. Having enough shift and rise to cover the negative.
2. The difficulty in aligning everything.
3. Sturdiness.

Cesar talked about using an 8x10 Sinar P (or similar). If you have one, and feel like giving it a try, then by all means do so. I've owned a couple of Sinar Ps, including and 8x10, and they are great cameras. My impression based on that experience is that they would be a PITA to use for this purpose. It's not something I'm going to pursue.

Cesar Barreto

20-Feb-2012, 11:25

Following the comments by Darrin and Peter, I'd say that using a Sinar or similar camera really can be time consuming but when I suggested 8x10' models I was considering bigger shift and rise movements.
And I always had some faith on Sinar cameras regarding alignment and sturdiness, so if one takes the obvious care, like mirror locking and so on, I don't see much to worry about.

I'm actually looking for an adapter wich allows to use a small camera right on the top of Sinar stand and I already missed three on the auction site. This way I could use a Nikon camera with it's own lens, but maybe one could make an adapter to fix the body to a lens board and so having the Sinar set with 3 stands: camera, lens and film.

One of the advantages of this solution is avoiding the use of glass, since the film is easily held flat when vertical, leading to less spoting and flare.

Peter J. De Smidt

20-Feb-2012, 12:39

I was a little curious as to why I needed fewer frames to cover the 6x6cm negative than I had anticipated. (I suspected my computational abilities.) So I photographed a mm ruler with my 55mm Nikkor. While focused at an indicated 1:1 on the lens, it covered a greater area than the size of the sensor. So I reversed the lens with an inexpensive F-mount to 52mm adapter, and the lens was still short. I added a Nikon K4 extension ring, though, at that brought me very close to a true 1:1. Hopefully, I'll have time to try a new scan tonight.

rdenney

20-Feb-2012, 18:00

While focused at an indicated 1:1 on the lens, it covered a greater area than the size of the sensor.

I thought the 55mm Micro-Nikkor required an extension tube or converter to achieve true 1:1, the same as the Canon 50mm f/2.5 Compact Macro.

Speaking of which, I actually forgot that I own that lens. It is extremely sharp with floating elements is well-corrected down to its minimum focus. But I lacked it's dedicated converter to achieve 1:1. KEH to the rescue--these seem to no longer be widely available and Canon may be giving up on this old, but excellent lens. If it works, it will be oh-so-much more convenient than the bellows arrangement.

Rick "whose project continuation today got interrupted by a wallpaper removal task, proving that visible spare time is a dangerous thing to display" Denney

Mike Anderson

20-Feb-2012, 18:05

I thought the 55mm Micro-Nikkor required an extension tube or converter to achieve true 1:1...
That's what I thought too. From this page (http://www.mir.com.my/rb/photography/companies/nikon/nikkoresources/micronikkor/55mmmicro.htm):

This highly rated lens can reach a maximum magnification ratio of 1:2 (0.5X) with its internal helical focus mechanism. With a matching PK-13 Auto Extension Ring, the ratio can reach 1:1 Life Size (1X)- so does when it is used in combination with a 2X teleconverter such as Ai-TC-200 or Ai-S TC-201. If you wish to achieve an even higher magnification, this lens is best to with a Bellow Unit such as Nikon PB-6 Bellow Unit. With such a possible combination, the magnification ratio can even be reaching to an astonishing 1:3.5 !

Peter J. De Smidt

20-Feb-2012, 19:03

There are two scales for the reproduction ratio engraved on the lens barrel, one with the lens alone and another when an extension tube is used. Apparently the model right before mine could reach 1:1 without an extension tube.

My earlier stitch was at 1:2, which only required 9 exposures for coverage and gave close to a 16x20 inch print @300 dpi. That's bigger than I printed medium format in the darkroom.

Daniel Moore

20-Feb-2012, 20:02

My 55 micro Nikkor requires an extension tube. I hadn't owned on prior to this project. I found the PK-13 extension tube for 35.00 shipped which seems to be about the going rate used.

Peter J. De Smidt

20-Feb-2012, 21:00

I expect it'll work best reversed. I bought a cheap f-mount to 52mm adapter new on Ebay for a few dollars. Daniel, do you have the F3.5 or F2.8 version?

el french

20-Feb-2012, 21:38

I don't think there's anything magical about 1:1 magnification. I would keep increasing the magnification until the lens can't resolve any more detail or you don't see any more detail in the printed image. I suspect the later will/should be the limiting factor.

Ben Syverson

20-Feb-2012, 21:58

1:1 is a pretty good goal for current lenses and cameras. The top DSLRs can resolve a real 60-70 lp/mm on the sensor with a decent macro lens, which should approximate the level of information in a color negative. In a few years, you'll be able to do better, but there's no reason to shoot for worse.

B&W of course is another issue. To get everything out of a fine B&W negative, you may need greater than 1:1 magnification, or a higher-resolution camera. We're probably a few years away from being able to do that easily.

Daniel Moore

20-Feb-2012, 22:38

Ben, I think we may be on that peak sooner than later. A D800E paired with the 55 Micro Nikkor or the 60mm f/2.8 might be all one would want or need, based on Peter's results so far. Peter, I've got the 2.8 version.

JimL

21-Feb-2012, 02:20

Did a little test of my own using my Pentax K-5 and a Pentax 645 120mm macro lens. The setup was rather crude: a dinky Manfrotto travel tripod with ball head sitting on my light table. The negative was taped to a piece of ground glass, and the negative was shifted under the lens using the small ruler to guide the shifting (overlap was done visually through the viewfinder). Magnification was about 1:1.5 and the 4x5 negative was shot in 4 rows of 5 frames. The .DNG files were color neutralized in Camera Raw, TIFFs exported, and the TIFFs stitched using Photoshop's Photomerge. My first couple of tries yielded some out of focus frames which was solved by putting more stable support under the tripod leg sitting on the light table's Plexiglas. The camera was leveled with a 2-way level sitting on the lcd screen, but there was a bit of geometric distortion which resulted in Photomerge making one edge of the negative a bit wonky. Tonally the results were identical, but the sharpness and detail from the DSLR stitch are impressive compared to the Epson V700 (2400 dpi scan). Now I'm curious to try at higher magnification. The Epson crop is unsharpened, but I just realized the Camera Raw frames would have been sharpened at CR's default settings. I didn't find shooting and stitching the 20 frames too onerous considering the results...

68662686636866168664

Peter J. De Smidt

21-Feb-2012, 05:32

Hi Jim. Nice test. Dslr scanning does seem to have a lot going for it.

SURF

21-Feb-2012, 05:38

Jim,
With a sharp applied to the Epson scan and balancing the sizes Pentax still feels like a better scanner than Epson.

buggz

21-Feb-2012, 07:00

This is an awesome thread w/ GREAT information!
I will have to try to rig up some things w/ my 5DMkII and bellows w/ the different macro lenses I have.
I have purchased the Epson V750M Pro as yet, but am still planning to do so.

buggz

21-Feb-2012, 07:56

Can one use a cheap Gagne Porta-Trace light box for this?
5K bulbs is what is wanted?
Yes, I am a newbie in this area, but really wanting to learn and experiment also.
Thanks!

Peter J. De Smidt

21-Feb-2012, 08:12

Sure, you can use a Porta-trace box. That's what I did some initial tests with. Mine is visually uneven, though. If you have a styrofoam box, take the top off. Put the box top down the on the light box, such that the open top of the box is sitting on the light box. Get a piece of diffusion material, I use rigid plastic, a little bigger than your negative. Cut a whole in the bottom of the box, which is now on top, a little smaller than your diffusion material. Tape diffusion material over hole. Turn on light box and see how even the source is. (You don't have to use a box. You could make one out of foam core, styrofoam sheets... It doesn't have to be very thick. 2 to 3 " should be enough.

rdenney

21-Feb-2012, 08:52

I don't think this will replace an Epson scanner, which is worth owning just for sheer convenience. The effort required to make this scan should be reserved for those negatives needing more enlargement than the Epson supports, or that benefit in some other way from using this method. The level of effort will be closer to what it takes to make a scan using a drum scanner. We won't achieve that quality, but it will be cheaper and it will be a usable method for much longer than drum scanners will be available and operational.

Summary: Even if my setup works as well as I hope it will, I'm not reserving the portion of my work area currently devoted to my Epson V750.

Rick "but the light table may have to find a new spot" Denney

Peter J. De Smidt

21-Feb-2012, 10:27

How long does it take to do a high resolution scan of a 4x5" negative on an Epson V750?

rdenney

21-Feb-2012, 11:15

How long does it take to do a high resolution scan of a 4x5" negative on an Epson V750?

I haven't timed it, but it's "a few minutes" rather than "nearly an hour". I don't know where this process will end up, but I suspect it will depend on the stitching software. Even then, I bet it's closer to the "nearly an hour" than the "few minutes". It would be less fiddly, I'm thinking, than a drum scan, but definitely more fiddly than an Epson.

Rick "who got redirected mid-project and now it may be a while" Denney

Peter J. De Smidt

21-Feb-2012, 12:48

Here's a test image that is much more challenging to stitch:

http://i955.photobucket.com/albums/ae37/peterdesmidt/Lighthouse_Stitch_Web.jpg

Again, it's a 6x7 Acros negative. It took 25 frames to cover it at 1:1 with my 55mm Micro Nikkor-P F3.5 reversed. This would make a print roughly 26 inches by 32 inches at 300 dpi.

Clearly, there are some stitching flaws, and there's a repetitive luminosity "motif" in the sky.

Here's a detail:

http://i955.photobucket.com/albums/ae37/peterdesmidt/Lighthouse_section_Web.jpg

If you'd like to work with the 8-bit tiff source images, you can download and extract the folder at:
http://dl.dropbox.com/u/3595413/Light%20house%20stitch%208%20bit.rar

Feel free to post versions here.

rdenney

21-Feb-2012, 12:59

Could the luminosity motif be a function of vignetting in the reversed lens?

Rick "thinking this might be a more common problem with shorter lenses" Denney

JimL

21-Feb-2012, 13:09

Jim,
With a sharp applied to the Epson scan and balancing the sizes Pentax still feels like a better scanner than Epson.

I made some test prints of crops at 400dpi, which would be a 32"x40" print and the results look crisper than 20"x24''s I've done from Epson scans, so I'm excited about this method. Considering the hours that go into Photoshop work on some images, the time and effort for the initial scan are acceptable to me. I can still see using the Epson for proofing and evaluation. The next thing to try is a color neg...

sethlatimer

21-Feb-2012, 14:36

Thanks so much for the lighthouse example...the reason I havent been working on this is I don't have a digital camera to take some pictures with so including the files is a godsend for me.

I am guessing that the repeating pattern in the sky is one of two things, uneven illumination from the light source or vignetting from the lens. In either case one solution is to take a picture of the light source without the negative on it and use its negative as a layer on each picture before merging. Just a thought. I guess I am assuming that the negative moved around a stationary light source instead of both picture and light source moving together.

I am excited to start working on merging these tonight after work.
Seth

Peter J. De Smidt

21-Feb-2012, 15:40

I've done a little testing, and I'm confident that my light source is the issue. It's not fixed down, and putting the flash in position can very easily cause the the whole source to slide. This gives a dark pattern very similar to what I see in the scan. I'll built a better light source soon. When I do, I'll upload more files for people to play with.

SURF

22-Feb-2012, 10:57

Hi Peter, I have some remarks. First, the shots are not that sharp. I am not sure in that, but it can be. Second, autostitching software can be confused with dark areas around the shots. It is better to shot without them. Looking forward for new files. Al

Darin Boville

23-Feb-2012, 00:04

A quickie test using Jim's source images. I simply ran these through Autopano Pro on my default settings. On my first try I had the strong banding/shadows as in Jim's test shot. Then I cropped the black areas out and tried again. Same result. Autopano didn't seem to care about the black areas. That was surprising. Then I vignetted a little to compensate in a rough guess sort of way for light fall-off in the lens. Stupidly I did this to the cropped images so not the best test (it's past midnight here, sigh). In any event, this seemed to lead to a noticeable improvement in the shadowy banding issue.

--Darin

68782

Daniel Moore

23-Feb-2012, 02:01

Autopano refuses to even add images to the group if it cannot find control points to align them with, thus the voids. Does anyone know a way to manually insert and roughly position featureless images into place within Autopano? PTGui does this easily. For all of Autopano's power I suspect I'm simply overlooking that option.

Peter J. De Smidt

23-Feb-2012, 07:13

Good morning,

As we know, this module of our project is focused on stitching. This, I expect, is the biggest hurdle we are faced with. If we can solve this, then we can worry about different lenses, focus stacking, various types of light sources... But it doesn't really pay to spend time on those issues if we can't solve this. I really appreciate all of the help I've received on this, as I'm a stitching novice.

With photos with details everywhere we've already had success, but the problem is images with large, homogenous and non-detailed areas, like those in the light house. It is not a good photograph, and I expect it isn't very sharp, as it was super windy that day, and I used a 10-stop neutral density filter. But for our current purposes, it seems like a good choice.

So far, I've had the best success with PTGui's align to grid feature, with the recurrent pattern in the sky being the main problem. This is most likely due to a light source issue. I'm almost done building a new light source, and it should be considerably more even. In the future I'll also include a picture of the light source, which might be usable to get rid of the variations from the pictures. I should have the new source down by mid-day, and when it's done, I'll shoot a new series of shots.

In addition to PTGui, I"ve dabbled a bit with Autopano Giga. Here are the results so far:

http://i955.photobucket.com/albums/ae37/peterdesmidt/Lighthouse_Autopano_Giga.jpg

Unfortunately, Giga doesn't allow me to use their align to grid process, as that seems limited to mosaics shot with the various motorized pano heads.

Would using something like Adobe® Lens Profile Creator be useful? With it we could minimize geometric and tonal distortions introduce by our lens before we stitch the individual frames. On this topic, does changing magnification change the distortion characteristics of a lens? The ALPC uses photographs of charts to make it's profile, and these would be easiest to make and photograph at non-macro magnifications.

I'm hoping that the very precise and repeatable negative stage movement offered by a motorized x-y stage will help with the stitching issue.

buggz

23-Feb-2012, 09:54

I've used Microsoft ICE(free BTW) on my 5DMkII stitches w/ very good success.
However, I have not tried it on stitches that have large expanses of openness, sky, water, etc...
I don't have any, have to generate some this weekend.
There is also another freeware stitching app, Hugin, I've used it, but it demands much more time and knowledge.
MS ICE was sooo much easier.
PS CS5 was so so to me.
Also, my experimenting of stitches were not related to this current project, I just learned of this project recently.

Darin Boville

23-Feb-2012, 09:56

>>A quickie test using Jim's source images.<<

Sorry! I meant Peter's source images....

--Darin

Daniel Moore

23-Feb-2012, 10:43

Lens correction is an unavoidable by product of all stitching programs. It's more a challenge to actually undo it's findings. As far as having different distortion characteristics at different magnification levels goes, I'm not sure but I don't think it's at all important to even try and create a profile with calibration software. My approach is to take an image with lots of detail and stitch it. Once the algorithm has aligned the mosaic I then check the results for the generated control points. I delete all of them with errors greater than 1 pixel. We know any such misalignment must be a fluke when stitching scans, or the film is buckled or something, which points to a problem other than parallax errors. Once the control points have been culled, re-optimize the image, don't create any new control points as you'll just be going backwards. If the preview looks good with the culled control points, it is good. At that point the lens will have been assigned a set of a, b, c distortion characteristics which will be all you'll need as far as calibration for future use. If you want to be really picky, push it further. Remove all control points greater than .7 and re-optimize. Then use the new lens distortion figures. These figures, at least in PTGui, can be saved as a preset. You now have a calbration for that magnification to assign (in PTGui it's 'load from database') future stitches.

Hats off to Helmut Dersch (http://webuser.fh-furtwangen.de/~dersch/), the father of PanoTools (http://webuser.fh-furtwangen.de/~dersch/), the progenitor of stitching programs.

Old-N-Feeble

23-Feb-2012, 12:41

Some interesting ideas here.

What if one were to... Remove the SLR lens from the equation by removing the CCD sensor from the camera body and "contact print" the film image directly onto the sensor with a well-regulated back-lit box then stitch those images together (no lens distortion). Even if the filtered glass becomes damaged those are replaceable, right?

What if one were to... Remove the scanner bar from a HQ flatbed and mount it in a fixed position, build a variable-speed-driven drum that's been precision smoothed and frosted and lit on the inside, and drive that system with the flatbed scanner's factory software?

alexn

23-Feb-2012, 13:33

Any half reasonable DSLR macro lens suitable for this kind of job will have next to no distortion anyway... I had a 105 F/2.8 VR Nikkor and now have the 200mm F/4 Nikkor and both are flawless for exactly this kind of job...

I have done a few slides with my D3X+200 F/4 Micro-Nikkor and got good results however it does not rival a good drum scan and to be honest, the results are not better than I have seen from a V700 or V750.

The scans I did were 6x17 slides at 1:1 with the 200mm F/4 on a rig I made up to position the camera above my light box and allow horizontal movement. once one row of images was taken the rig was moved down 20mm and the next row was shot.

3x6 35mm images with reasonable overlap resulted in a 390 odd megapixel final image.

My light box is an electroluminecent sheet under a piece of clear perspex.. I take an image first of the light box to gain the correct exposure. I then use this image as a custom white balance to be sure that the colours in the slide are correct.

I then lay my slide ontop of the piece of perspex, lay another 2mm sheet of anti-reflection glass ontop of the slide to ensure flatness and begin imaging. It took perhaps 1.5 hours to get the rig set up, expose, white balance correct, then image the 18 sections. A further hour to stitch and clean up (used ptGUI and have used Photoshop's Automate -> Photomerge in the past.) Another 30~40 minutes to clean up any dust marks (both from the slide, and dust marks from the D3X's sensor... So we are up to 3 hours to get a lesser result than you can achieve with a V700 in 30 minutes for 1/10th of the price of my D3 and 200 F/4... Oh, and the V700 does not require you to make a custom rig to do the scan, does not require you to be bent over taking the images and setting it all up for a 3 hour period. you load your slide/neg and let it rip...

Forgive me for being negative on the matter but the old saying "if it aint broke, dont fix it" comes to mind.

Flatbeds work fine and if you want higher quality, get a good drum scan done..

In AUD, my D3X cost $8900, the 200 F/4 was $1400, the lightbox setup including perspex and glass plates cost about $150 and the rig I made costs about $250.

For just shy of $10,000 AUD you can get a hell of a lot of drum scans done.. Or you could buy a ton of film, and a V750 and start doing what we all love doing. Making photographs.

Peter J. De Smidt

23-Feb-2012, 14:22

I'm not going to waste time arguing about the feasibility of the project for the umpteenth time.

The second lighting prototype is done. Before making it, I spent some time looking at my De Vere mixing box, and I learned a few things. (I had based my first light source off of the mixing box on my Philips PCs2000.) First, there's not a 45* panel opposite the light source. Instead, the "reflector" panel is at a much smaller incline, just enough that on the end opposite the light source the reflector comes up exactly the to the level of the opening through which the light enters the box. The idea, I expect, is to spread the light from the source over as much of the reflector as possible. The De Vere uses a profiled piece of diffusion plastic to even out the light. Instead of that, about 1 inch above the high point of the reflector sheet, I put a sheet of diffusion. Another inch up (or so) there's the final sheet of diffusion plastic. The lighted opening is bigger than before, and I've fixed everything in place, so that the whole apparatus can't shift when I attach the flash.