Copy Move Image Forgery Detection Using Keypoint Based Approach

Abstract

Identifying and detecting portion that has been manipulated in an image is a challenging research area. Therefore, in the given paper, we propose a fusion for copy-move forgery area detection based on identifying Scale Invariant Features in an image without using any reference image. Here, features are extracted using SIFT algorithm and matched using Brut force matcher. The identical portions are clustered using BIRCH clustering to detect the forged portions in an image. In our study, we considered natural images which are tampered using image manipulating tools. The experiments are performed on publicly available datasets Viz., MICC-F220, MICC-F600 and results obtained are compared with other existing methods in the literature along with provided ground truth images.