Robust image hashing based on structural and perceptual features for authentication of color images

Image hashing is one of the most celebrated techniques regarding the discipline of image forensics, image retrieval, image indexing, content verification, and zero watermarking. For such sensitive and complex problems, generation of a unique and robust image hash is an utmost prerequisite for an image identifier driven from the perceptual contents of an image. As a design perspective, it is essential for an image hash to have robustness and optimized discriminative capability. We propose a robust image hashing technique by acquiring perceptual features based on a novel distance magnitude profile utilizing color pixel incongruity among the contiguous pixels, as well as producing a structural image for discrimination. The combination of both results provide a state-of-the-art robust profile that has efficient discriminative capacity and is impervious to usual signal processing violations.