High performance optical image hiding based on computational ghost imaging with BCH error correction coding technique

Abstract

The conventional image hiding scheme based on computational ghost imaging (CGI) has low-quality reconstructed watermark and unsatisfied security. In this paper, a high performance optical image hiding scheme is proposed to cope with these issues. The watermark is first read as a binary number sequence and processed by exclusive-OR operation with a random number sequence that generated by chaotic system. Thereafter, the result is coded with Bose-Chaudhuri-Hocquenghem error correction code and encrypted by the CGI system. Finally, the obtained cyphertext is quantified into two levels and hidden in the host image. Differing from other watermarking schemes, binary images, grayscale images and color images can be used as watermarks to be reconstructed losslessly, and a series of simulation results illustrate the feasibility and effectiveness of the proposed scheme.