Copy-move operation is an essential block in most of digital image forgeries. It results in hiding information from a scene, inserting new information or simply duplicating it. Many studies have been made in order to detect this kind of image modifications. In this paper we propose an algorithm that applies principal component analysis and discrete cosine transform in order to characterize and detect duplicated blocks in an image. The algorithm is optimized and tested on a database of forged images. Obtained results compared to a reference method proves the adaptability and performance of the algorithm.