Geometric Invariant Robust Image Hashing Via Zernike Moment

Robust image hashing methods require the robustness to content preserving processing and geometric transform. Zernike moment is a local image feature descriptor whose magnitude components are rotationally invariant and most suitable for image hashing application. In this paper, we proposed Geometric invariant robust image hashing via zernike momment. Normalized zernike moments of an image are used as the intermediate hash. Rotation invariance is achieved by taking the magnitude of the zernike moments. Image normalization method is used for scale and translation invariance. A randomization diffusion processing enhance hashing security. The test results show that our method is robust with respect to the geometrical distortions and content preserving processing.