In our society digital images are a powerful and widely used communication medium. They have an important impact on communication and IT industry. The proposed versatile over division calculation sections the host picture into no overlapping and sporadic blocks adaptively. Then, the element focuses are removed from each block as block elements, and the block components are coordinated with each other to find the named highlight focuses; this technique can around show the presumed forgery districts. In past few years, research goes to detecting and classified for copy move forgery images for forensic requirement. So detection is very important challenges for testing in forensic science. In this paper detection and classification by point base and block base features SIFT and SURF Respectively but use ant colony optimization in matching and feature selection phases ,in case of SIFT features and proposed SIFT with ACO features which also use in classification with support vector machine with Gaussian and polynomial kernel