Robust Detection of Copy-Move Forgery Using Texture Features

Cloning or copy-move is a special type of forgery that tries to hide the regions of the image by a block that is copied from the same image. In this paper we introduce a new texturebased method to detect such forgeries in digital images. In the proposed approach, at first the image is divided into some overlapping blocks and then we utilize modified Gabor filter to extract the feature vector of each block. In the next step, to reduce the dimension of the extracted feature vectors, PCA algorithm is applied on them. Finally, in the matching step, we use counting bloom filter to determine similar or duplicated blocks. The experimental results show that the proposed features are very effective in accurate detection of copied regions, even when these regions have undergone lossy compression. The accuracy and performance of our method in comparison to similar works, proves efficiency of the proposed theory and shows noticeable increase in detection rate

Keywords:

digital forensics , forgery detection , cloning , Gabor filter , PCA algorithm