A Main Object-oriented Projective Invariant Image Watermarking Approach

Abstract

Digital Watermarking is a technology that enables to protect distributed digital content from unauthorized reuse. Robust watermarks must be able to survive a range of geometrical "attacks," such as image resizing, cropping, translation, rotation, scaling, smoothing, filtering, and simplification. In the current paper we extend existing approaches that are based partially on image invariant features; by introducing new image invariant features and exclusively using them. The proposed watermarking method is based on a combination of image processing techniques, theory of geometry, multiple image invariant features, and public key encryption. Experimental results show that the use of such combination of techniques allowed for securing the embedded watermark and for success in image watermark's recovery, if images were exposed to several geometrical attacks.