Discrete cosine transform features matching-based forgery mask detection for copy-move forged images

In this age of digital technology, it is relatively simple to manipulate images. It is quite difficult to recognize altered photographs with the human eye, and as a result, we are frequently fooled or misled on social media platforms as a result of altered images being shared. As a result, it is of the utmost need to work on methods that are both quicker and lighter in order to categorize and pinpoint tampered or faked images. In this article, we demonstrated a methodology for detecting copy-move forging masks that uses a modified version of the discrete cosine transform (DCT) block level features extraction method. The altered photos have their feature information retrieved at the block level. a procedure in which blocks of the proper size are chosen experimentally. In addition, these block characteristics are compared by measuring the lexicographic distance between the feature vectors in order to find blocks that are similar to one another. The visual findings are offered for comparison between the real mask and the one that was anticipated. The detection accuracy of fake masks is around 95% on average.