Pareidolia, but underwater: What is under the green paint?

[Jorge_Stolfi]

The main problem I have with automated pareidolia is the same I have with the natural one - the texture of the vellum itself provides a lot of subtle lines and curves, with enough squinting or tweaking of the parameters of models one can see/reveal almost anything.

Indeed.  But this approach is at honest at least in the sense that the final classification is made independently for each pixel, based only on its color; without trying to look for multi-pixel patterns like lines or characters.  Which is where actual pareidolia comes in.  It will be left to the human user to "see" such patterns on the computed probability maps.  The user's choice of sample pixels will influence the classification, but only through their colors, not through their positions or adjacency relations.

I suppose if a model is trained on this folio it's possible to get much more curios results, but I was actually aiming for maximum certainty, so I only trained on samples from 6-7 folios and then applied the result to the whole MS.

I think that training separately for each page (or even for each clip from each page) is justified because there seems to be some overall variation from folio to folio in the colors of parchment, ink, and paint.  For instance, the green paint of You are not allowed to view links. Register or Login to view. seems to be more bluish than that used in the Bio section.  

And there is also a much larger variety of paints and stains over the whole MS.  On this particular image I don't have to include the red, blue, yellow, and rusty paints as separate provinces. They will be classified as "OTHER" without significantly affecting the classification of the other provinces of interest.  And I don't have to worry about ketchup stains, or the gray offset from blue flowers, that are important "noise" features on some other pages.

All the best, --stolfi