Passive Copy-Move Image Forgery Detection Method Using MSER and SVD

Digital Forensic is a branch of forensic science which is related to computer crime. It deals with the investigation and recovery of material found in digital devices. Digital images and videos are the most important part of digital forensics as they are the prime evidences in law issues or in mass media. So the loyalty of the digital image is important. Due to the rapid increase and easy access of photo editing software tools, it has become easy to duplicate and modify digital image. Copy-move forgery (CMF) is the widely recognized image forgery technique in digital images. In CMF, certain part of the image has been copied and subsequently pasted to another location within the original image to make duplication or cover something. A considerable number of researches already done on the copy-move forgery detection (CMFD) but there’s a problem occurs in feature extraction and matching phase. So in the proposed work, we design copy-move forgery detection system using the MSER feature extraction technique and correlation based matching on the feature vector. MSER is the application for the region presented in the image so we can find out the forged region easily. Experimental analysis shows that the proposed work achieves better accuracy on the basis of precision and recall rate compared to similar works.