Abstract: Using smart phone and digital technology it’s easy to recapture images and videos from an liquid crystal display (LCD) monitor and from smart phone. Such image and video recapturing activities are a security threat, which allows the forgery images/videos to bypass the current forensic systems. The task of verifying the ownership and history of an image or video is, consequently, more difficult. One approach to detecting an image or video that has been recaptured from an LCD monitor or from smart phone is to search for the presence of aliasing due to the sampling of the monitor or screen pixel grid. In this paper, we show that it is possible to detect recaptured image and video based on feature set of image. By using LCD monitors or smart phone, its add aliasing effect in image, so using features like brightness, contrastness, edge width, blurriness etc., we can identify recaptured image. We find the fact that the edge profiles of single and recaptured images are marked different and we train two alternative dictionaries using the KSVD approach. One dictionary is trained to provide a sparse representation of single captured edges and a second for recaptured edges. Using these two learned dictionaries, we can detect whether a query image has been recaptured with use of Support Vector Machine (SVM). Experiments conducted show that the proposed algorithm is capable of detecting recaptured images and video with a high level of accuracy.

Keywords: Smart Phone, LCD, Aliasing, Edge Profiles, K-SVD, Sparse Representation, SVM.