Digital Image Forgery Detection using Graph Fourier Transform

Abstract

This paper proposes a novel method using graph based features for detecting copy move forgery done in digital images. For extracting the features of the image patches the graph Fourier transform are used. A graph can represent data in different applications with a generic data structure. A graph different from classical transforms like DCT and DWT in the sense that it can represent signals on irregular and high dimensional domain. A graph can fit to local characteristics through edges for regular signals like images. Graph Fourier transform extracts the frequency interpretation for signals on graph. To reduce the processing time of feature matching patch match algorithm is used which is a fast approximate nearest neighbor search algorithm. The experiments conducted on the GRIP database shows that the proposed method is highly accurate and are robust to rotation and scaling.