Perceptual Image Hashing
Cluster focuses on perceptual hashing techniques for detecting visually similar images despite modifications like cropping, compression, or pixel changes, with frequent mentions of phash.org and PhotoDNA.
Activity Over Time
Top Contributors
Keywords
Sample Comments
That's perceptual hashing. Check out https://www.phash.org/
Or perhaps it uses file hashes to determine if certain images are similar.
curious question from a non-programmer - are you checking against the exact same image (i.e. hashed), or is there an easy way of trying to match an image to a very similar one you've seen before?
Yep. Google "perceptual hash functions".
Perceptual hashes are very good for that, maybe with some adjustments for mirrored images and some crops
This seems like an almost perfect use case for one-way image hashing.
furthermore they are perceptual hashes. its not like you can just defeat it by changing a pixel in all your images
You are confusing image recognition with classification. They use recognition, similar to reverse image search services. Not really 'data-hash comparison', but rather 'image-hash', since it may see through re-compression and other minor modifications.
It finds specific images, but the hash is based on pixels, not raw file bytes. The hash is insensitive to small image changes (brightness, saturation, rotation, compression artifacts), so slightly modified images still hash to the same value.
PhotoDNA relies on perceptual hashing?