A copy-move forgery is introduced in images by copying a part of an image and putting it within the same image or in the other image. Recently, copy-move forgery detection has been actively considered by many researchers. The detection of copy-move forgery for edited image is very tedious to find, because, many persons use the copyrighted image of other for their commercial use without the permission of owner. Even if the owner finds the similarity, there is no particular method existing to check if the reference image has been used in the target image. In order to solve this issue, a key point extraction technique is proposed in this paper to detect the similar points in the reference and target images. This paper presents an algorithm for extracting key points as Scalable Invariant Feature Transform (SIFT) and Speeded-Up Robust Feature (SURF) features based on the pixel-by-pixel correspondences between the reference and the target image. Consequently, an affine transform in the matrix form is estimated and refined by using Expectation-Maximization algorithm. Finally, the existence of copy-move forgery is confirmed by the affine transform matrix estimated. The SIFT and SURF features proposed to be used in this work are robust to different noise sources and reliable for the identification of copied regions. © 2016, International Journal of Pharmacy and Technology. All rights reserved.