Novel zero-watermarking method using the compressed sensing significant feature

In this paper, we proposed a novel zero- watermarking method using the compressed sensing (CS) significant feature. The original image is segmented into non-overlapping blocks and then constructs a scrambled block Hadamard ensemble (SBHE) matrix as a sensing matrix to sense every block. CS significant feature can be represented by the sum of each block’s measurements. Then each pixel value of the binary zero-watermark is embedded by modifying the CS significant features. At last zero-watermark is registered, and the sensing matrix is stored as extra key. With the CS significant feature, the robustness of the zero-watermarking algorithm can be greatly improved. SBHE matrix can reduce the computational complexity of the compressed sensing and also reduce storage costs. In addition, because of the randomness of SBHE matrix, the proposed zero-watermarking method can provide excellent confidentiality. Moreover, our algorithm is different from the traditional zero-watermarking methods with pure meaningless image features, copyright holders can be legally recognized by the identity’s image which is embedded into the zero-watermark, so the authentication process of the zero-watermark is more intuitive and convenient.