Attacking Some Perceptual Image Hash Algorithms

Perceptual hashing is an emerging solution for multimedia content authentication. Due to their robustness, such techniques might not work well when malicious attack is perceptually insignificant. We designed an experiment and verified that some state-of-the-art image hash algorithms could not distinguish small malicious distortion and some authentic distortion. We proposed an enhancement framework as a remedy. It suggests extracting information from the content and combining it with the secret key to generate the perceptual hash, so that perceptually insignificant information can be protected.