Bandelet Union Optimal Matrix Norms Constructing Lossless Watermarking

Abstract

A multi-frequency lossless watermarking algorithm is proposed in this paper. Bandelet and optimal matrix norms are used to draw the geometric feature and the statistic feature. The bandelet transform is applied to the image, and the texture and edge traced by using the geometric flow via feature of the image to build the high frequency lossless watermarking. A low frequency watermarking scheme is also achieved, by selecting optimal matrix norm to process the low frequency part of wavelets coefficients. The experiments show the offered watermarking system is evaluated against standard watermarking attacks.