Robust Image Hashing Based on Multiple Histograms

Image hashing is a novel technology of multimedia and finds many applications such as image retrieval, image copy detection, digital watermarking and image indexing. This paper proposes a multiple histograms based image hashing, which can reach an acceptable trade-off between rotation robustness and discrimination. The proposed hashing is done by converting the input image into a normalized image, dividing it into different rings, extracting ring-based histograms and compressing them by discrete wavelet transform. Hash similarity is evaluated by L2 norm. Experiments show that our hashing is robust against content-preserving manipulations such as JPEG compression, watermark embedding, scaling, rotation, brightness and contrast adjustment, gamma correction and Gaussian low-pass filtering. Receiver operating characteristics (ROC) curve comparisons indicate that our hashing has better performances than two existing algorithms in classification between perceptual robustness and discriminative capability.