会议专题

The Digital Watermarking Technology based on Neural Networks

This paper concerns the digital watermarking technology on three different processing stages in order to enhance the robustness of digital watermarking under the premise of invisibility. (1) The first stage is watermark signal pre-processing. The watermark signal created using binary gray images is taken the highly nonlinear processing by the chaotic function first and then by the neural network, which enhances the degree of watermark confidentiality greatly; (2) I he second stage is watermark embedding strength degree. First a neural network is constructed and trained. The trained neural network can be used in watermark embedding and extraction by which a watermark algorithm can be achieved to do blind detection; (3)The third stage is watermark embedding and extraction. The treated watermark signal is embedded into the airspace of the original image through the trained neural network. And the neural network is also used to extract and detect the watermark. Proved by experimental results, this algorithm has good robustness.

component airspace embedding strength preprocessing neural networks

YU Chang-hui FENG Wan-li ZHOU Hong

Faculty of Computer Engineering Huaiyin Institute of Technology Huaian, China

国际会议

2011 IEEE 2nd International Conference on Computing,Control and Industrial Engineering(CCIE 2011)(第二届计算、控制与工业工程国际会议)

武汉

英文

5-8

2011-08-20(万方平台首次上网日期,不代表论文的发表时间)