(10-12-2025, 01:02 AM)Jorge_Stolfi wrote: You are not allowed to view links. Register or Login to view.PS. But please pick a page...
Let's try f8r, I think it's a fair choice - it has quite a lot of variation in the density of the ink, not very damaged surface and at the same time there are MSIs available. For me personally it's one of the "sleeper" pages, I don't think I have spent any significant time on it, so it would be useful for me to have a better look in any case.
Edit: if I was doing this and checking if my method worked, I probably would first investigate the page based on the normal visible light TIFFs, identify and finalize the list of retracing candidates and only then check them on MSIs. Just to find out how many of the retracings from the normal scans would still look plausible on MSIs and how many won't.

