Copy-Move forgery detection and classification using SRVM

Digital images and videos are of great importance. However, the relative ease of manipulation of images and videos makes their authenticity unreliable. Among the types of image forgery, copy-move forgery is the most common model. In this type of image forgery, a part of an image is copied and pasted in another part of the same image. Now, various block and key methods have been introduced to copy-move forgery detection. The proposed method is a fake image classification method with the help of a combined SIFT and RVM model, named SRVM. In the proposed method, features of the object are analyzed. Then based on the training data, the kernel function is obtained and finally the classification process will be performed. The IMD and MICC datasets were examined for the accuracy of copy-move forgery detection. The results show $81 %$ accuracy for IMD and 9 4 % for MICC.