Detection of Image Region Duplication Forgery Using Model with Circle Block

Due to the widespread use of sophisticated imaging tools, digital image forgeries have already become a serious social problem. In this paper, we proposes an efficient and robust algorithm for detection of a specific category of digital image forgery known as region duplication forgery, which is done by copying a block of an image and pasting it on to some other block of the same image. The image is first reduced in dimension by Gaussian pyramid, and four features are adopted for each circle block. The feature vectors are lexicographically sorted. Then similar vectors will be matched by a certain threshold value. Finally, the area threshold value is proposed to remove the wrong similar blocks. Experimental results show that our method is robust to not only the post processing including noise adding, blurring, lossy compression, but also the copy region rotation. Meanwhile, the proposed method reduces the total number of blocks to narrow block-matching searching space, which can improve the efficiency comparing with the exiting methods.