An Improved Perceptual Hashing Algorithm of Image Based on Discrete Cosine Transform and Speeded Up Robust Features
Xiaojian Ding · Journal of Chinese Computer Systems · 2014
In order to solve the problem of poor robustness of global features and high complexity of local features in image perceptual hashing algorithm,the author proposed an improved perceptual hashing algorithm based on DCT and SURF. Using DCT as global features and SURF as local features,the author gave the hashing functions and the fusion of two features. Then,the author provided the application in image authentication. Experimental results showed that this algorithm has good robustness and efficiency.