RGB digital image forgery detection using Singular Value Decomposition and One Dimensional Cellular Automata

A reliable approach for RGB digital image forgery detection is presented in this paper. Our method is based on the Singular Value Decomposition (SVD) of the original image in which the features of the image is extracted and then pushed into cellular automata to generate the Robust Secret Key for the image authentication. SVD is used as a strong mathematical tool to decompose the RGB Digital Image into three orthogonal matrices and create features that are rotation invariant. First we focus on a specific layer of RGB input image (i.e. red layer or red matrix) and apply SVD to calculate the Singular Values as well as Right and Left Singular Vectors. Then we generate an array to hold the values of the Singular Values and the Right and Left Singular Vectors and then push the values of the array into the One-Dimensional Cellular Automata to generate the Robust Secret key that can be used to protect the image against the forgery. Next, we embed the output of the cellular automata into the spatial domain of another layer, to authenticate the original image. We applied our algorithm on hundreds of 800 × 800, RGB digital images to test the behavior of our algorithm. The experimental results have illustrated the robustness, visual quality and reliability of our proposed algorithm.