A zero-watermark system based on combined spatial and wavelet domains

To protect the copyright of digital images efficiently, while not degrade its quality, this paper proposed a dual-domain zero-watermark algorithm, which increases the flexibility and robustness of single domain method. This algorithm embeds the zero-watermark in the highest bit planes of spatial domain and wavelet domains' low frequency sub-image simultaneously, based on content's importance. Experimental results show that this method has good safety and strong robustness. However, all zero-watermark algorithms produce large sideband information which brings forth great challenges to the application of zero-watermark. Hence the paper analyzes the sideband information produced in embedding process and points out that different keys have different significance, which provide the theory bases to key compression and grade management.