Copy-Move Forgery Detection and Performance Analysis of Feature Detectors

Now a days, the digital image integrity are remarkably important for the exchange of data which are generally utilized for different applications like fraud detection, therapeutic imaging, reporting, and advanced crime investigation. Digital images can easily be forged with the advancement of image manipulation tools and information technology. The commonly used image forgery technique in digital forensic filed is Copy-move forgery. The two fundamental classifications for identifying copy-move forged images are keypoint-based and block- based method. Block-based strategies have the burden of high computational expense because of the enormous number of image blocks and it fails to deal with different geometric transformations. On the contrary, keypoint-based methodologies can overwhelmed these two draw-backs however are discovered hard to manage smooth locales. As a result, these two methodologies are combined and proposed a effective copy-move forgery detection. Also, we accomplish a comparative study between different keypoint detectors and feature matching algorithms used to determine computational complexity of each.