Paper:
Language: English:

Intelligent Logo Watermarking Based on Independent Component Analysis

Thai Duy Hien^, Zensho Nakao^, and Yen-Wei Chen^,*

Faculty of Engineering, University of the Ryukyus, Okinawa 903-0213, Japan
College of Information Science and Engineering, Ritsumeikan University, Shiga 525-8577, Japan
Institute for Computational Science and Engineering, Ocean University of China, China

Received: September 25, 2003

Accepted: March 29, 2004

Keywords: digital watermarking, independent component analysis, robustness

Journal ref: Journal of Advanced Computational Intelligence and Intelligent Informatics, Vol.8, No.4 pp. 390-396, 2004

Abstract

We present new intelligent logo watermarking based on independent component analysis (ICA) in which a binary logo watermark is embedded in a host image in a wavelet domain. To improve robustness, an image adaptive watermarking algorithm is applied by a stochastic approach based on a noise visibility function (NVF). The algorithm design, evaluation, and experimentation are described. Experimental results show that the logo watermark is perfectly extracted by ICA with excellent invisibility and with robustness against various image and digital processing operators and almost all compression algorithms such as Jpeg, jpeg 2000, SPIHT, EZW, and principal component analysis (PCA) based compression.

[Notice]
* "Preview" is the first 2 pages of the article. You don't need the registration.
* To read the PDF file you will then need to download and install the Adobe Reader.
Adobe Reader is free and available for download here: