Robust image forgery localization and recognition in copy-move using bag of features and SVM

Nowadays in the world of advanced digital computer, tampering and implication of digital images can be easily performed by a tyro with a number of accessible advanced image processing softwares like Adobe Photoshop, Corel Draw etc. Therefore it becomes very challenging for end users to distinguish whether the image is novel or forged. In the fields such as forensics, medical imaging, e-commerce, and industrial photography, legitimacy and integrity of digital images is crucial. This motivates the need for detection tools that are transparent to tampering and can reveal whether an image has been counterfeit just by scrutinizing the counterfeit image. Recently many new methods are presented with improved performances of detection. However there is still place to improve this performance further. Some of the proposed state-of-the-art image tamper detection techniques have been selected for proposal. In the training stage, it extracts the keypoints for every training image using the Mirror invariance feature transform (MIFT), a vector quantization technique maps keypoints from every training image into a unified dimensional histogram vector (bag-of-words) after K-means clustering. This histogram is treated as an input vector for a multiclass SVM to build the training classifier. In the testing stage, the keypoints are extracted and fed into the cluster model to map them into a bag-of-words vector, which is finally fed into the multiclass SVM training classifier to recognize the tampered region.