- Notifications
You must be signed in to change notification settings - Fork2
Help find similar duplicate images for the Internet Archive
License
NotificationsYou must be signed in to change notification settings
jpypi/dup-image-search
Folders and files
| Name | Name | Last commit message | Last commit date | |
|---|---|---|---|---|
Repository files navigation
This project is to help the Internet Archive find duplicate images for their many images (particularly music album art covers).
Thanks tohttp://www.hackerfactor.com/blog/index.php?/archives/432-Looks-Like-It.html for the Simple Hash and pHash (aka perceptive hash, which uses DCT)
- MD5 checksum (This may change to SHA.)
- Simple Hash
Scale to 8x8, greyscale, hash based on above/below average - DCT (Discrete Cosine Tranform)
Scale to 32x32, greyscale, [DCT](https://en.wikipedia.org/wiki/Discrete_cosine_transform), hash based on above/below average excluding top-left "base" value