I posted the question yesterday and it seems obviously that some people are interested in given the count of upvotes. But it does not seem to be very useful, as some Mathematica guru puts it. However, I think the reverse question is quite useful, i.e., how to reconstruct the original image in Mma. One answer given by a user actually used image processing software and did a decent job on reconstructing the original picture. It would be great to know how to split on tiles and use some feature matching. in Mma automatically.
Thank you all.
Halfway there, posting on this old question just to see if someone wants to give it a try:
l = Import@"https://i.sstatic.net/R3xte.jpg";
l1 = Binarize[GradientFilter[MeanShiftFilter[l, 3, .01, MaxIterations -> 10], 3]
// ImageAdjust, .1];
l2 = Erosion[Dilation[l1, 20], 20];
l3 = ImageData@ColorNegate@l2;
n = 1; While[l3[[1 ;; n, 1 ;; n]] == Array[1 &, {n, n}], n++];
ip = ImagePartition[l, n][[1, 1]];
ipMsft = MeanShiftFilter[ip, 3, .1, MaxIterations -> 100];
l8 = ColorNegate@GradientFilter[ ImageMultiply[ip, ipMsft], 6] // ImageAdjust;
ImageSubtract[l8, GradientFilter[l8, 1]]

If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With