A Rotationally Invariant Texture Descriptor to Detect Copy Move Forgery in Medical Images

The wide spread use of multimedia communication and advancement in image processing techniques are the key factors that makes forgery easy. The copy-move is a very common forgery in digital images. Most of the techniques detect a copy move forgery of size less than 16×16. This paper, presents an efficient method to detect copy move forgery detection in medical images using center symmetric local binary pattern (CSLBP) which is able to detect the forgery size up to 12×12. The proposed block based method is robust against geometric distortion, gaussian blurring, JPEG compression and additive white gaussian noise. Simulation results exhibit that the proposed method outperforms many other well-known methods.