A robust high-capacity watermarking algorithm

We present a new robust high-capacity watermarking algorithm using the block-based chaotic map and the intersection-based pixels collection. The block-based chaotic map is created to break the local spatial similarity of an image partially. The intersection-based pixels collection is utilized to generate a container and a reference register which is used to indicate the location of the container for watermark embedding and extracting. Experimental results demonstrate that our algorithm can achieve robust high-capacity performance in comparisons with the current watermarking algorithm. It can embed more than 5,050 bits of information in an image of size 202X202. It can achieve a low message error rate of less than 0.22 even under various attacks such as additive noise, low-pass filtering, and lossy compression.