基於互補式總體經驗模態分解的音訊浮水印技術

碩士 === 國立高雄第一科技大學 === 電腦與通訊工程研究所 === 104 === In recent years, along with the upgrading of internet speed and the popularity of mobile devices and personal computers, the circulation market of digital multimedia is booming, so how to protect the copyright of the publisher became fairly important. The...

Full description

Bibliographic Details
Main Authors: Meng-Chen Chou, 周孟辰
Other Authors: Wen-hising Lai
Format: Others
Language:zh-TW
Published: 2016
Online Access:http://ndltd.ncl.edu.tw/handle/16623680891716915843
Description
Summary:碩士 === 國立高雄第一科技大學 === 電腦與通訊工程研究所 === 104 === In recent years, along with the upgrading of internet speed and the popularity of mobile devices and personal computers, the circulation market of digital multimedia is booming, so how to protect the copyright of the publisher became fairly important. The digital watermarking technology can provide appropriate protection for the copyright, and the ownership of multimedia information is easy to confirm. It is quite effective method for copyright protection. In this thesis, we propose to use complementary ensemble empirical mode decomposition replace empirical mode decomposition in audio watermarking technique to improve ability of against external attack, and combine group of samples modulation to solve the problem that quantization index modulation using in time domain can not resist the volume scaling attacks. We also use BCH coding and Arnold transform to improve robustness and security of the audio watermarking technique. In experiment part, we use four kinds of music to verify the effectiveness of audio watermarking technology. The results show that we can extract the clear and recognizable watermark image after attacked by noise addition, volume scaling, resampling, requantization, MP3 compression or echo addition, and the watermarked audio signal still maintain the acceptable quality.