In recent years, digital images are used in a wide range of applications and for multiple purposes. They additionally assume an essential job in the capacity and exchange of visual data, particularly the mystery ones. With this far-reaching utilization of computerized pictures, notwithstanding the expanding number of instruments and programming of advanced pictures altering, it has turned out to be anything but difficult to control and change the real data of the picture. In this way, it has turned out to be important to check the validness and the respectability of the picture by utilizing current and computerized procedures, which add to examination and comprehension of the pictures substance, and after that ensure their honesty. There are numerous kinds of picture fraud, the most imperative and well-known sort is called duplicate move imitation, which utilizes a similar picture during the time spent phony. This kind of imitation is utilized for one of two things, first to shroud an item or scene by duplicating the zone of the picture and sticking it on another zone of a similar picture. The second is the redundancy of item or scene with a change in a few characteristics, "for example, measure" by duplicating this article and sticking it on another zone of a similar picture. In the proposed work, we have implemented the Modified SIFT (M-SIFT) algorithm along with multi SVM classifier technique to detect the copy-move forgery in the digital images.