Description

An effective method for the detection of forgery in images is proposed. In recent years, digital images are in use in a wide range of applications and for multiple purposes. They also play an important role in the storage and transfer of visual information, especially the secret ones. With this widespread usage of digital images, in addition to the increasing number of tools and software of digital images editing, it has become easy to manipulate and change the actual information of the image. In our proposed system, we are going to propose a method for copy move forgery detection which can sustained various pre-processing attacks using a combination of Dyadic Wavelet Transform (DyWT) and Scale Invariant Feature Transform (SIFT) with BFSN(Broad First Search Neighbours) clustering and CFA(Colour Filter Array) features. In this process first DyWT (Dyadic Wavelet Transform) is applied on a given image to decompose it into four parts LL, LH, HL, and HH. Since LL part contains most of the information, we intended to apply SIFT on LL part which combines Broad First Search Neighbours (BFSN) clustering and Colour Filter Array (CFA) features. SIFT extract Key features and then we make BFSN clusters of those key features and then we perform cluster matching and CFA features. BFSN clustering algorithm is applied to detect multiple copied areas in tampered images. In order to distinguish original regions from tampered regions, we take CFA features into consideration to discover the inconsistency at the edges of the copied regions. And by using DyWT with SIFT which combines Broad First Search Neighbours (BFSN) clustering and Colour Filter Array (CFA) features we are able to extract more numbers of key points that are matched through BFSN clusters and thus able to detect copy-move forgery more efficiently. From the identified similarity between the features the copy move regions were identified in the image. The performance of the process is measured with the help of precision, recall and F-measure of the input image.