Perceptual hashing library

 Unlike cryptographic hash functions which rely on an avalanche effect of
 changes in input leading to drastic changes in the output, perceptual
 hashes are "close" to one another if the inputs are visually or
 auditorily similar.
 .
 Perceptual hashes have applications in searching, identifying potential
 copyright infringement, locating duplicates etc. - since a fingerprint
 of a file is usually less than 1% of the original media file's size,
 the fingerprint can be stored in a database and used for semantic
 searching, such as finding all images that look like another.
 .
 pHash implements several perceptual hashing algorithms and provides a
 C++ API to use those functions in your own programs.