Content Based Invariant Image Watermarking with High Capacity

Abstract

Recently, many content based image watermarking schemes have been addressed to resist geometric attacks, such as rotation, scaling and translation (RST). The underlying idea of this kind of method is to align the geometric transforms before watermark embedding and extraction. These schemes are usually preceded by watermark synchronization, an operation to determine some local regions for watermark embedding and extraction. Due to the small size of the regions, the capacities of these schemes are usually limited, typically tens of bits. In this chapter, we propose two methods that can embed high capacity watermark images, while maintaining robustness to the general geometric attacks. Compared to the existing schemes, the proposed high capacity watermarking methods embed the watermark in a content based manner. The efficiencies of the proposed schemes are demonstrated by experiments.