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

[oshfdk]

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.

Yes, this was my reasoning too for only using local models. Also, it helps if the result can be independently reproduced via a simple (ideally linear, or maybe polynomial) combination of channels. 

For example, I still think that the strongest argument that the signature like feature at the bottom right of f116v actually exists is that it can be reproduced by a linear combination of MSI channels and it's possible to verify this with a simple Python script: You are not allowed to view links. Register or Login to view.

The more complex processing is required, the less persuasive the result is, because I think for chaotic processes it's possible to keep tweaking the training set or the training parameters to highlight random combinations of pixels to produce the result one desires.