Improving Invisibility and Robustness for DCT-Based Image Watermarking against JPEG Compression

碩士 === 國立臺灣大學 === 電機工程學研究所 === 99 === Digital watermarking is the process of embedding information into owner’s contents that is difficult to remove. It can represent in several ways such as an embedded tag, picture, signature or any digital information. Once a digital content has been stolen, owner...

Full description

Bibliographic Details
Main Authors: Haw-Shiang Chang, 張浩祥
Other Authors: 雷欽隆
Format: Others
Language:en_US
Published: 2011
Online Access:http://ndltd.ncl.edu.tw/handle/71176478498357590463
Description
Summary:碩士 === 國立臺灣大學 === 電機工程學研究所 === 99 === Digital watermarking is the process of embedding information into owner’s contents that is difficult to remove. It can represent in several ways such as an embedded tag, picture, signature or any digital information. Once a digital content has been stolen, owner could use digital watermark to claim the copyright of ownership and apply for compensation according to copyright law. To improve the robustness of digital watermark, most techniques proposed to use discrete cosine transform (DCT) to embed watermark information into the low- frequency part of image since DC components have larger perceptual capacity than AC components. These schemes perform very well under a general JPEG compres- sion. However, when the watermarked images have been compressed with higher compression ratio, the embedded watermark is destroyed. In this thesis, we proposed an innovative DCT-based digital image watermarking scheme that could enhance the robustness under higher JPEG compression with the concept of modulus. We could also preserve the visual quality of watermarked images.