Image hashing is a short and compact sequence, which can represent image itself. It is an important technology in the field of multimedia information security. At present, image hashing has been widely used in image authentication, image indexing, copy detection, image quality assessment and image watermarking. In this paper, we propose a novel image hashing with visual map and SVD (singular value decomposition). A contribution is feature extraction from image visual map. Since visual attention model can well retain visual region information, desirable discrimination of hashing algorithm is achieved. In addition, hash generation via SVD from visual map ensures good classification performance between robustness and discrimination. The effectiveness of hashing algorithm is validated by two open databases.

Full text: IJEIR_2919_FINAL.pdf