Slide holding for Scanning - Ohnar and Other Options

Bernard Delley wrote:

your metal object got you near a serious MTF blade test:

You could use a backlit razor blade as target, and then analyze a section of the blade edge using mtf_mapper. Considering the tiny DoF, you should move the target in ~ 0.1 mm steps to find optimal sharpness after analysis.

Such tests were discussed in an earlier thread.

Thanks. I've not come across those tests before, nor that term. If I understand correctly it is testing for edge sharpness. My metal shapes are chemically etched from 10 thou material but imperfections in the etching process mean that they are unlikely to be as 'clean' as a razor blade.

My main concern though is the bleeding of lighter parts of the image into dark areas.

It occurred to me today that there are another two variables involved - the monitor and the observer. I can't do much about the second variable.

I made an SVG simulation with completely solid grey areas of similar luminance, exported it to PNG so that it was 'like for like' and compared this with the image from the camera.

I see similar bleeding when looking at the SVG derived image on screen so there is an issue here with the monitor or the observer.

Does that mean that the digitised slides don't really have this issue? I don't think it does. If I stretch tones in the relevant dark areas (both of the slides and the metal shape test photo) with curves then I still see some halo which is definitely from the image.

It does seem possible that the monitor is making my perception of the problem worse than it really is. I'm not sure that I can do much about the halo (or whatever the right term is) in the captured image anyway - switching between lenses doesn't seem to improve things but maybe there is an optical improvement that I have not come across. From what I've read there aren't many easy options to remove it in postprocessing either.