Copy-move forgery detection in images via 2D-Fourier Transform

Digital images have been widely used in many applications. However, digital image forgery has already become a serious problem due to the rapid development of powerful image editing software. One of the most commonly used forgery techniques is Copy-move forgery that copies a region of an image and pastes it on the other region in the same image. In recent years, most techniques aim to detect such tampering. Different feature extraction methods have been used to improve the capability of the detection algorithm. In this work, we used two dimensional Fourier Transform (2D-FT) to extract some features from the blocks. Predetermined number of Fourier coefficients hold information about the blocks. At the final stage, the similarity search between the adjacent feature vectors is performed to determine the forgery. Experimental results show that proposed method can detect the duplicated regions with high accuracy rate even if the image is distorted with blurring mask or it is compressed with different JPEG quality factors. The dimension of feature vector is also lower than the other methods in the literature. Thus, the method ensures the lower feature vector with high accuracy rates. The proposed method also detects multiple copy move forgery as shown in the results.