Abstract

The key to watermarking algorithm is to extract good watermarking information on the basic of watermarking invisibility. Though many watermarking algorithms have been proposed till now, the watermark information only is disturbed by simple permutation. Also, the character of those permutations is too evident, the invert transformations are easy as regularities, parameter of transformations is less. In this paper, a novel watermarking approach based on Discrete Wavelet Transform(DWT) and Fresnel diffraction transforms has been presented. Firstly, the watermark image is transformed to binary kinoform by using Fresnel diffraction transforms. Then, using DWT, the original image is divided into different classes, the binary kinoform is embedded in the third level sub-band by the rule of odd-even adjudgement. The extracting algorithm can extract the binary kinoform and invert transform to watermark image without the original carrier image and the original watermark image. The experimental results show that the method is robust against some image processing such as JPEG lossy compression, median filtering, Gaussian smoothing and random cropping attacks and so on. Experimental results show also that watermark image via Fresnel diffraction transforms has good concealment performance, and robustness of watermark is increased.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call