Apply the Counterpropagation Neural Network to Digital Image Watermarking

• Main Authors: Sheng-Jyun Su, 蘇聖鈞
• Other Authors: Chuan-Yu Chang
• Format: Others
• Language: en_US
• Published: 2005
• Online Access: http://ndltd.ncl.edu.tw/handle/88379658557651764606

碩士 === 國立雲林科技大學 === 電子與資訊工程研究所 === 93 === The rapid development of computer network and multimedia technology makes it easier to assess digital media. Since the problem of illegal reproduction and modification has become more serious than before. Digital watermarks are an important technique for protection and identification that allows authentic watermarks to be hidden in multimedia. In this thesis, we propose a novel method called Full Counterpropagation Neural Network (FCPN) for digital image watermarking, in which the watermark is embedded and extracted through specific FCPN. Different from the traditional methods, the different watermarks are embedded in the synapses of a FCPN instead of the cover image. Therefore, the watermarked image is almost the same as the original cover image. In addition, most of the attacks could not degrade the quality of the extracted watermark image. Moreover, the watermark embedding procedure and extracting procedure is integrated into the proposed FCPN and it also accomplishes watermark embedding and extraction in one watermark by multi-cover image or multi- watermark by multi-cover image. The experimental results show that the proposed method is able to achieve robustness, imperceptibility and authenticity in watermarking.