Perceptual image hashing using block truncation coding and local binary pattern

In this paper, we propose a novel image hashing scheme based on block truncation coding (BTC) and local binary pattern (LBP), which can be applied in image authentication and retrieval. In the proposed scheme, the pre-processing is first conducted on input image by bilinear interpolation, Gaussian low pass filtering, and singular value decomposition (SVD) to construct a secondary image for regularization. Then, BTC is applied on the secondary image to obtain the high/low quantized levels and the corresponding binary map that can reflect the contents of image. The concatenated image feature sequence is generated with the assist of the center-symmetrical local binary pattern (CSLBP). Finally, data dimensionality reduction is exploited on the image feature sequence to produce the part of hash. Combined with high/low quantized-levels, the final hash can be obtained. Experimental results show that the proposed scheme has the satisfactory performances of robustness, anti-collision, and security.