Improved Copy move forgery detection approach-A Comparative study [articol]

Abstract

In recent years, digital images are in use in a wide range of applications and for multiple purposes. There are many types of image forgery, the most important and popular type is called copy move forgery, which uses the same image in the process of forgery. Copy–Move and cut-paste forgeries are common type of tampering, where part of an image is copied and pasted to another place to add a new object or to hide/remove an object already present. Most of the copy move forgery detection methods are interest point-based, where the significant key points are extracted and compared to each other to locate similar regions and another category of method called block-based, where the image is divided into overlapping blocks and then features are extracted and compared to find similar regions. In this paper a new hybrid approach is proposed that exploits the block based and the key point based approach for copy-move forgery detection hence it is considered as “hybrid” method. In this paper two approaches have been presented, one with regular regions and another one with irregular image regions for forensic detection of copy move forger and their performance is analyzed and reported. In the first approach, the key points are extracted from the image, and a set of connected triangles are built onto these points to model the foreground regions. Then the triangle region matching is done using their inner angles, color information and area of the triangle regions to locate the copy move tampered regions. In the second approach an adaptive over segmentation method is employed so that the input image is divided into non overlapping regions of irregular shape and key points are extracted from the resultant segmented irregular blocks. For similarity matching of regions a method based on correlation is employed. A comparative study was made by conducting experiments on MICC F220 and MICC F2000 databases.