Copy-move forgery detection based on hybrid features

Copy-move forgery is one of the most popular tampering artifacts in digital images. Most existing keypoint-based detection methods may fail to precisely locate duplicated regions because there are not enough correct matched points. In this paper, a novel copy-move forgery detection method is proposed based on hybrid features. A robust interest point detector KAZE is introduced and combined with SIFT to extract more feature points. In order to deal with multiple duplications, an improved matching algorithm is used which can find the n-best matched features. Then an effective filtering step based on image segmentation is executed to filter out false matches. Moreover, an iteration strategy is developed to estimate transformation matrices and determine the existence of forgery. Based on these matrices, the duplicated regions can be located at pixel level. Experimental results demonstrated that the proposed method can precisely detect duplicated regions even after distortions such as rotation, scaling, JPEG compression and adding noise.