Detect duplicated pages by visually comparing them

[stefnotch]

Someone finally sent me some PDFs that have duplicated pages where the pages metadata got lost.

Here, the best way of identifying duplicates would probably be:

• Comparing text (easy one, me thinks)
• Comparing the visual output, and preferably checking if a lot of pixels have either become darker (usual slides: white background, dark foreground) or lighter (dark theme slides). This is rather slow. (Use a library like https://github.com/mapbox/pixelmatch )

[stefnotch]

Test_fur_PDF_Stripper.pdf

[stefnotch]

More test files algorithmics_ws22_part1-1.pdf

algorithmics_ws22_part2-1.pdf

algorithmics_ws22_part3-1.pdf