Mean-Classified and Hierarchical Block Matching Watermarking in Fractal Coding

This study presents two watermarking approaches in fractal coding images. The first method, namely mean-classified fractal coding (MCFC), partitions the domain blocks of an image into groups according the mean values. As documented in experimental results, a high embedding capacity can be achieved with good image quality. Moreover, the MCFC is proved of highly robust to against many types of attacks, including JPEG, tampering, and rotating. The second method, namely hierarchical block matching fractal coding (HBMFC), utilizes different domain block sizes to enlarge the searching pool with slightly sacrificing the coding gain. The HBMFC is mainly adapted to fragile watermarking, which can be applied to authentication application to prevent from illegal alteration.