Angle Quantization Index Modulation Based on Block Compressive Sensing for Robust and Secure Image Watermarking

摘要

A novel angle quantization index modulation watermarking scheme based on block compressive sensing (BCS-AQIM) is proposed to embed robust and secure watermark. In this algorithm, sparse random matrices constructed by chaotic sequence are chosen as the measurement matrix and the measurement vector of host image is obtained by block compressive sensing. In each block, two optimal measurements are chosen to embed the watermark into their angle. Then, the watermarked image is recovered through reconstruction algorithm. The measurement matrix and the position of watermarked measurements are used as a key. The performance of BCS-AQIM under AWGN attacks and its security are analyzed and assessed by simulations. Experiment results demonstrate that the proposed method is robust to different types of attacks and outperforms common existing methods in terms of the robustness.