Abstract
For a secure watermarking scheme, the watermark embedded in the original image should be perceptually invisible and robust against various attacks. The Lin's watermarking scheme embeds watermark bits in the original image by adjusting the luminance of the pixels around the edges of the features. Depending on Kerckhoff's principle, the embedding and extraction algorithms are finally known to the public. Unfortunately, the watermark embedded with the Lin's scheme is vulnerable once an attacker knows the information about the embedding and extraction algorithms. In this paper, we propose a method to improve the security of the Lin's scheme. Instead of being embedded in the designated blocks around the edges of the features, the watermark bits are embedded in the randomly selected blocks in the original image. The fixed modulus used to control the strength of the watermark in the Lin's scheme is adaptive in our proposed scheme to gain the robustness and transparency. Experiment results show that the security of the Lin's scheme is efficiently improved by our proposed method. Moreover, the robustness property of the Lin's scheme still remains in our proposed method.
Original language | English |
---|---|
Pages (from-to) | 11-22 |
Number of pages | 12 |
Journal | Chung Cheng Ling Hsueh Pao/Journal of Chung Cheng Institute of Technology |
Volume | 31 |
Issue number | 2 |
State | Published - 05 2003 |
Externally published | Yes |
Keywords
- Copyright protection
- Digital watermarking
- Just-noticeable-distortion
- Stegoanalysis