Copy-move forgery is a type of image that is most commonly used. This technique is easy to use by many people. This, application can be used to detect copy-move forgery in digital image. The application convert the input image from RGB to graysacle. Then apply Discreate Cosine Transform method to perform the image decomposition, and to extract the image features with the SIFT. Then feature extraction results are clustered using nearest neigbour and estimated by geometric transformation using RANSAC method. SIFT local features is better for some geometric transformation, such as rotation and scaling which used in this study. Result showed that this application is able to detect copy-move forgery in digital images with or without rotation and scaling attacks up to 100% with threshold in 0.10.