Rotation Invariant Copy-Move Image Forgery Detection Using a Method Based on Fuzzy-Zernike Moments

Prabhjot Kaur; Khyati Marwaha

Abstract

AbstractWithout leaving any visual clue, digital images are easy to manipulate due to the availability of countless editing software packages. Any image manipulation can become a forgery. Copymove forgery is a very common method of creating forge images by copying some part of the image and pastes it on the same image. In this paper, we propose a copy-move digital image forgery detection method using Fuzzy-Zernike Moments. The proposed method combines the fuzzy logic with Zernike Moments for better detection rates. The goal of this paper is toimprove the accuracy in the detection of forged area. The image is first resized and divided into overlapping blocks. Then the features of the blocks are extracted using Fuzzy-Zernike Moments technique. After, matching is performed by using lexicographic sorting. Finally, filtering is performed to reduce the probability of false matches by using correlation technique. Experimental result shows that the proposed method can detect duplicate regions efficiently.Keywords: Digital Forgery, Authenticity, Copy-Move Forgery Detection, Image Forensics, Zernike Moments, Fuzzy LogicCite this ArticlePrabhjot Kaur, Khyati Marwaha. Rotation Invariant Copy-Move Image Forgery Detection Using a Method Based on Fuzzy-Zernike Moments. Journal of Computer Technology & Applications. 2016; 7(2): 58–68p.