LA2's profile - activity

I tried this and it looks promising. The example has the drawback that it only looks for the best match, even if that match is rather poor (for example, in a page without portraits). I settled for CV_TM_SQDIFF_NORMED and skipped the normalization. Then I tried to pick out the best candidates. However, my medium-grey mask missed some of the portraits (that were dark grey) and instead suggested hits in the text of the page. I think the page image needs preprocessing for this approach to be successful. Perhaps erosion can make the text go away.

Not an answer, just a minor correction to your illustration: Once you have found Y1 and Y2, the formula (Y2-Y1)/2 gives you the distance between Y1 and Y3. The actual position of Y3 is that distance + Y1, which means that Y3=(Y2+Y1)/2.

It would be really useful if someone would take the time to compile a translation dictionary between the terminology used in GIMP and/or ImageMagick (and Photoshop and Netpbm, for that matter) and the one used in OpenCV, as they sometimes differ.

Just to mention one example, in GIMP you can "shrink" or "grow" a selection, but a search on the OpenCV website for the word "shrink" gives no hits, while "grow" gives many false leads. In OpenCV (and computer vision textbooks) these concepts correspond to algorithms for "eroding and dilating", which are not words that a newcomer will search for. Perhaps that OpenCV tutorial page should mention "in GIMP, this concept is called to shrink and grow a selection", or the translation dictionary should be a separate page. Either way would be useful.

How can I automatically detect, crop and save the oval shaped portraits in scanned book pages like this one? http://runeberg.org/spg/17/0183.html I have 20,000 portraits in 6000 pages in this format to begin with. After that I have more books where the shapes might be slightly different. (This is not face recognition. I just want to find the oval shapes.) This question was also asked on the Wikimedia Commons graphics forum, https://commons.wikimedia.org/wiki/Commons%3aGraphics_village_pump#Pulling_illustrations_from_scanned_book_pages

A different example, still with oval portraits, but with a different page layout and no contours, is here, http://runeberg.org/pgsmal/0141.html