Digital Watermarking in Image Affine Co-Variant Regions

In this paper, we present a robust approach of digital watermarking embedding and retrieval for digital images. This new approach works in special domain and it has two major steps: (1) to extract affine co-variant regions, and (2) to embed watermarks with a novel fuzzy inference mechanism. The advantage of this proposed approach is demonstrated by a large amount of experiments, which are mainly against different attacks, such as common image processing, rotation, scaling, cropping, and several other affine transformations.