Image Forgery Detection using Adaptive Oversegementation and Key Point Matching

Abstract

Abstract: The scope of this paper is to propose a image forgery detection application by combining adaptive oversegmentation and feature point matching, the suggested scheme presents a novel method for detecting copy-move forgery in images. Using superpixel-based analysis, it divides the image into erratic blocks, extracts feature points, and fine-tunes areas that could be fakes. The accuracy of the detected forged areas is then improved by morphological operations. Results from experiments indicate that this method performs better than current ones, especially under difficult circumstances.