Block-Based Forgery Detection Using Global and Local Features

Abstract

Nowadays, many image-editing tools have emerged. So image authentication has become an emergency issue in the digital world, since images can be easily tampered. Image hash functions are one of the efficient methods used for detecting this type of tampering. Image hashing is a technique that extracts a short sequence from the image that represents the content of the image and thus can be used for image authentication. This method proposes an image hash that is formed using both the global and local features of the image. The Haralick texture features are used as the local feature. The global features are based on the Zernike moments of the luminance and the chrominance component. This robust hashing scheme can detect image forgery such as insertion and deletion of the objects. The features are extracted from the blocks of the image and so can detect forgery in small areas of the image also. The proposed hash is robust to common content-preserving modifications and sensitive to malicious manipulations.