Abstract: Copy and Move is a very common way of image tampering. In order to detect images through rotation, scaling and other operations quickly and efficiently, this paper presents a way of blind image tamper detection based on Radon and Fourier-Mellin transform. In the first the article blocks the image, then it calculates the Radon and analytical Fourier-Mellin transform of the image block and extracts the characteristic value of transformation results, and finally it calculates the value of the correlation moment features. This algorithm does not require gray image binarization and normalization, and it directly form the graphics Radon transform and Fourier-Mellin transform to extract invariant features of the results. The theoretical analysis and experimental results show that the proposed algorithm result is better than detection method based on orthogonal moments, and the white noise with mean zero was significantly higher than the robustness of the detection method based on orthogonal moments.