Chinese Journal of Computers   Full Text
Title	The Research of Border Problems’ Influence on Watermarking Algorithms Based on Wavelet Transform
Authors	WANG Gang RAO Ni-Ni
Address	(School of Life Science and Technology, University of Electronic Science and Technology of China, Chengdu 610054)
Year	2007
Issue	No.10(1872—1879)
Abstract &   Background	Abstract Among the watermarking algorithms, the ones based on the Discrete-Time Wavelet Transform(DTWT) are widely used. However, the border problems in DTWT make it more difficult to extract the watermark for the watermarking algorithms. After studying the border problems for watermarking algorithm based on DTWT, the authors propose that the watermark had better not be embedded in the border coefficients of the sub-band, and give out the proof theoretically. Experimental results show the validity of the theorems. Then the authors applied the theorems to three watermarking algorithms. Simulations show its superiority is over the schemes neglecting the border problems. keywords discrete-time wavelet transform; border problems; watermarking algorithm; border data; border coefficient background The work is supported by the Science and Technology Key Projection of Ministry of Education(No.03146). In this project, digital watermarking is the main research focus, which is also a key problem in the current field of telemedicine. The research outcomes of this project will provide copyright protection or content authentication of medical images. The research group has proposed some novel fragile and robust watermarking algorithms for medical images of telemedicine based on DTWT. It is widely accepted that robust watermarking algorithms should largely exploit the characteristics of the HVS(Human Visual System), for more effectively hiding a robust watermark. A widely used technique exhibiting a strong similarity to the way the HVS(Human Visual System) processes images is DTWT. As a matter of fact, the next-generation image code standard JPEG2000 strongly relies on DTWT for obtaining good quality images at low coding rates. So watermarking algorithms based on DTWT play a key role. In DTWT,the image has only a finite number of samples,and if the length of wavelet filter is more than 2, we have to extend the original image.This is a so-called border problem. The border problems’ influence makes watermarking algorithms more difficult to extract the watermark. After studying the border problems for watermarking algorithm based on DTWT, the authors give the definition of border data and border coefficients for DTWT, propose that the watermark had better not be embedded in the border coefficients of the corresponding sub-band, and give out the proof theoretically in this paper.