Improvement in Copy -Move Forgery Detection Using Hybrid Approach

Abstract

In this present digital world, digital pictures and videos are the main sources of information. However, these carriers of information can be easily tampered by using softwares such as Adobe photoshop, GIMP etc. Thus, the issue of verification of authenticity and integrity of digital images becomes necessary. Copy Move Forgery is a popular type of forgery that is commonly used for the manipulation of digital images. In this, a region of digital image is copied and then pasted to another location with in the same image with intension to make an object disappear from an image by covering it with small block copied from another part of the same image. There are several post processing operations that are applied by manipulators to obstruct the forgery detection techniques. Thus, for aforementioned problem, we in this paper proposed a method which is a combination of SIFT and SURF algorithms. In this firstly image is split in to sub-parts by DWT method and then SIFT and SURF are applied to actual components of image one by one. After this, features extracted by both methods are matched to locate the forged part in the image. The experiment shows that the proposed method is more efficient and provides better results than applying SIFT and SURF alone.