An Efficient Technique for Passive Image Forgery Detection Using Computational Intelligence

This chapter proposes a scheme for the identification of copy-move forgery by inducing adaptive over-segmentation and matching of feature points with the help of discrete cosine transform (DCT). Copy-move forging is an image tampering technique that involves concealing undesired things or recreating desirable elements within the same image to create modified tampered images. Traditional methods added a large number of false matches. To conquer this problem, a new algorithm is proposed to incorporate an adaptive threshold method. So, the block feature matching mechanism is used, and the matching feature blocks classify the feature points using patch matching and Hough transform. Forged regions are detected with the help of the newly proposed algorithm. The results of the proposed method show that it can substantially reduce the number of false matches that lead to improvements in both performance and computational costs. This demonstrates the suggested algorithm's resistance against a variety of known attacks. Comparative results are presented for a better evaluation.