Title :
A Watermarking Method Using Quantization and Statistical Characteristics of Wavelet Transform
Author :
Byun, KeunDuck ; Lee, Sangjin ; Kim, HyungJun
Author_Institution :
Korea University, Seoul, Korea
Abstract :
A statistical watermarking method is proposed in this paper. The proposed method uses two dzferent algorithm according to the statistical characteristic of wavelet subbands of m input image. We embed a logo watermark at the middle-frequency band for the robustness of it. The proposed method is a blind scheme, so we need only the watermarked image when the embedded watermark is extracted. We achieved PSNR values around 40dB as shown in the experiment results. It has vely fast processing speed and also robust against attacks.
Keywords :
Data mining; Discrete cosine transforms; Discrete wavelet transforms; Image coding; Information security; Protection; Quantization; Robustness; Watermarking; Wavelet transforms;
Conference_Titel :
Parallel and Distributed Computing, Applications and Technologies, 2005. PDCAT 2005. Sixth International Conference on
Print_ISBN :
0-7695-2405-2
DOI :
10.1109/PDCAT.2005.57
