Based on the temporal correlation of video frames copy-move tampering detection method

The present invention provides a video frame copy-move tamper detection method based on spatio-temporal correlation, which is first downscaled video frames according to the method of establishing the Gaussian pyramid images, video content reuse monitoring time correlation and space correlation each set of adaptive time threshold and the peak threshold, after filtering and phase correlation exists suspicious intra frame copy-move tampering may be detected based on a block matching method, and calculates a binary diagram for describing the tampered position . The last frame of suspicious packets using the time a final decision threshold. The present invention is greater than similar algorithm reduction in computation time, and to enhance the robustness of the algorithm is the Gaussian noise. When the binary image using the tampered described, morphological filtering operations can be added to eliminate the interference noise false point and point to bring a certain extent, to improve the detection accuracy.