gridlike patterns in my scanned images....

I have been scanning 4 x 5 b/w negatives with my epson 3200 run by twain software, imported by Photoshop 7. The images are scanned at 360 dpi's. When i have the images imported in Photoshop, i can see the edges of the subject with jagged outlines almost as if they were pixellated. If i use the zoom tool the effect disappear, but if i magnify the image even more i can see some kind of gridlike patterns, throughout the image. The pattern is formed by lines of pixels (about 6 pixels wide) intersecting at 90 degrees with otherlines of pixels. I don't think these are Moire' because i am talking about Negatives scans. Does anybody know or suspects the culprit of this ? I am growing increasingly frustrated with the problems that softwares and computers keep entertaining me with: the all 2001 Space Odissey is coming back to me,

It's hard to tell. Maybee if you post the image it will make it easier to help.

Since you mentioned that the patterns dissapear when you zoom in a bit, I wonder if they might be caused by the software which has to render the image on screen. If you are viewing the image at an odd magnifacation, like 43%, then the software can pixelate things a bit.

Another possibility is JPG compression. JPG's produce some of the ugliest images I have ever seen. The compression takes nice gentle gradient tones and turns them into big ugly blocks of color. Try scanning as a tiff and see if the patterns are still there.

Moire occurs on a screen when the pixels of the image interacts with the dots that makes up the image on the screen. So it's got nothing to do with if you scan a negative or a positive. Since it comes and goes depending on at witch size you view it, my guess is some kind of moire. It could very well be enhanced my JPG compression as Darin suggested.

Perhaps scanning at 360 ppi is contributing to the problem. I suggest using some simple fraction of 3200, the smaller the better. If 3200 ppi produces files that are too big, try 1600 ppi or even 800 ppi.

I agree that it is an artifact of the display rather than something in the actual scanned image.

If the jaggedness changes alternately with zooming in then what you are seeing is the consequence of the zoom factor mismatching the screen spi (resolution). Photoshop is not WYSIWYG in that regard.

If you are seeing jaggies upon extreme zoom, then its natural. You really do have jaggies. 360 is not a fine scan. For 8x10 enlargements you must double that. And soforth on up. The 3200 does not perform very well at 1200 in my opinion.

Domenico, do you have your monitor set to "millions of colors?" If it is set to "thousands" or "hundreds" of colors, you will see a grid on your images that is not in the scans, it is just in the way the monitor is displaying them. Go into your control panel and make sure the monitor is set to the highest possible number of colors; that might fix your problem.

~cj

Hi chris, the monitor was set already to millions of colours. But i tried ;to enhance the monitor resolution, and other than everything getting smaller and more defined, the grid problem is still there. Leonard, the scans are at 360 dpi, but the target size is 8 x 10 , these scans are for publishing purposes, not printing on paper. The final scans in CMYK are around 35-40 mb each. Could the gridlike pattern be some kind of electrical interference ? I read somwhere sometime ago that it can happen. Now that i think about it , i Had scanned some images yesterday mornng and they were fine , only the ones scanned in the evening had the problem . I have a table lamp plugged in the same outlet where the scanner is : i will try to scan with the light off and see what happen, if that works i will start laughing and never stop.

Electrical interference can cause grid or line patterns to appear on scans - you just need a periodic noise signal that modulates the pixel values in a regular way. Plugging in a lamp shouldn't of itself cause a problem, but the ballasts in some low-wattage fluorescents can radiate badly, and some power supplies can inject a lot of fuzz onto the ground wire. Also, the lamp or its cable may be changing the local strength of signals from radio stations or, say, wireless broadband links so that they start to interfere. Consumer equipment should be properly shielded, but it's often not if you look closely enough.

Posting an actual-pixels portion of an image would help diagnosis a lot. Failing that, it's worth calculating the frequency of the noise so you can start guessing where it's coming from. Interference patterns tend to either stay the same (at actual pixel magnification) or precisely double or halve in size if you double or halve the scan rate, so another clue can be had by seeing how the grid changes as you change the scan resolution.