Audio watermarking

• 主要作者: Hertanto, Richard Nathaniel
• 其他作者: Foo Say Wei
• 格式: Final Year Project
• 語言: English
• 出版: 2015
• 主題: DRNTU::Engineering::Electrical and electronic engineering
• 在線閱讀: http://hdl.handle.net/10356/65304
• 機構: Nanyang Technological University

Rapid and widespread copying of copyrighted audio data has become a problem in this current age due to the ease of doing it. Usually, it is impossible to determine the originator of the illegal copies of a file. Some schemes have been developed to address this issue; one way is by inaudibly encoding each legal copy with a unique signature (a watermark) and thus, correct identification of the signature can aid in finding the person responsible. This project attempted to improve an existing algorithm by modifying the way watermarks are encoded and decoded. This was achieved by capitalizing the fact that stereo system has 2 channels: instead of marking only one channel or both channels similarly, the scheme modifies them in different but related way. The 2nd channel serves as reference and helps to determine the actual value of the 1st channel. This scheme also allows the recovery process to return a decoded watermark not only whether it has binary value of 0 or 1 but also whether they were conclusively so and thus can aid in preventing false identification.