Victor,
You’re better off removing at the scanning stage if you can risk wetting the print…..
The method I have used with great success is as follows:
1) Place a thin sheet of glass on the scanner bed – picture-frame glass will do. This will stop the scanner bed getting messy.
2) Coat the print in glycerine solution – it will "fill-in" all the texture.
3) Scan
4) Wash the print in luke-warm water and dry flat
I don’t know what scanner/TWAIN you have. My HP Precision Scan Pro has a option "De-screen Printed Original" under "tools". Perhaps your scanner/TWAIN program has something similar. Mine isn’t foolproof and takes a bit of fussing with it but it does a fairly good job of getting rid of the half tone screen pattern.
there’s a descreen topic in the faq (yea, who reads that?!! :))
Mathias Vejerslev "De-Screening in Photoshop" 10/19/01 4:23am </cgi-bin/webx?13/0>
Let me amplify my comment of "takes a bit of fussing". About all that you can change is the DPI scan resolution. And it can make a difference (usually smaller is better). Keep in mind that WYSIWYG doesn’t hold true. You may or may not see interference patterns that are really not there (caused by your display screen resolution). Print it out to check what you really have.
Len’s method is messy. A better way (have not checked it out but it should work) would be to use a sheet of glass, plastic or even a pile of transparent sheet protectors between the object and the scanner bed to move it slightly out of focus. Because of it’s higher index of refraction, plastic would be better than glass. You just want to blur or destroy the half tone dot image properties.
About all that you can change is the DPI scan resolution.
I also seem to remember something about rotating the actual picture/photo to be scanned at different angles on the scanner bed to defeat the pattern.
Dave,
That can help with a regular halftone screen, but paper matting/texturing is rather more random and wouldn’t benefit from rotations.
ah. thanks for the clarification len. i thought it was you who posted that tip originally but i didn’t want to misquote you so i didn’t attrib it. and i wasn’t sure if it applied here or not, but I wanted to make sure all the bases were covered.
thanks, dave
"I read that it had some serious side effects. I can’t remember any of them, or else I could be informative. But at least I can be alarmist."
– Dave Foley
In the link given in post #3, LenHewitt wrote:
"Scan the halftoned image at a resolution to yield approximately 4 pixels per halftone cell. (IOW, if the image is screened at 150 lpi, scan at 600 dpi.)"
How do I know the screened lpi on a halftoned image – whether it is 133, 150 or 200 lpi, etc? Do I measure the number of colour "rosette" patterns per scanned inch on screen?
Or does it not matter too much to bother?
Mike
You can get a tool to determine the halftone screen on a printed image for $10 to $20 at a graphic arts supply store. (If it is worth it to you.)
Is it a half tone print, a texture pattern, like on silk finish prints, or a product of the age of the print, ie foxing, where the dye has come off and the scanner is illuminating the remaining silver base? The latter is not regular and any filtering will just knock out sharpness. Sometimes the longest route is the best and you have to clone it out. Another way is to duplicate the layer, zap a filter to remove the problem on the upper layer and then use the eraser to put holes in the upper layer so that the important sharp areas on the layer below show through. These will then be repaired with hands on one-by-one removal of the remaining dots.
Hope it helps.
John
