Image Recapture Detection Using Multiple Features

With advances in image display technology, recapturing good-quality images from highfidelity artificial scenery on a LCD screen becomes possible. Forgers can recapture the artificially generated scenery and use the recaptured image to fool image forensic system. Image recapture detection is to distinguish real-scene images from the recaptured ones. An image recapture detection method based on multiple feature descriptors is proposed in this paper, which uses combinations of low-level features including texture, noise, difference histogram and color information. One hundred and thirty-six dimensions of features are extracted to train a support vector machine classifier with RBF kernel. Experimental results show that the proposed method is efficient with good detection rate of distinguishing realscene images from the recaptured ones. It also possesses low dimensional features and low time complexity.