Spatio-temporal digital video hashing using edge orientation histogram and discrete cosine transform

In this paper we propose a spatio-temporal digital video hashing scheme. In the proposed scheme, the digital video is treated as a three dimensional signal. Edge Orientation Histogram (EOH) is computed for each frame in the digital video, then a temporal discrete cosine transform (DCT) for the magnitude of each EOH bin is computed. A subset of the resulting DCT coefficients are then pseudo-randomly rearranged to construct the hash. Our experiments show that the proposed system has good discriminating power, is robust against content-preserving attacks but is sensitive to content-altering attacks.