Image perceptual hash algorithm based on target character

This paper proposed a perceptual hashing algorithm based on the image shape target Using the Hu invariant moments, the scheme extract the image characters. Then, calculate the weighted average of the Hu invariant moments of the image and the normal processed image. Select five invariant moments which are good at differentiating normal image processing from malicious attacks to the hash. In authentication period, calculate the distance of two hash values to judge two images whether match or not. Finally, experimental simulations and theoretical analysis are used to demonstrate the effectiveness of this algorithm.