Ica And Its Applied Research In Digital Watermarking

Independent Component Analysis (ICA) is a novel method of signal processing and data analysis. Using ICA, the source signals can be extracted only from the observations according to the statistic independent characteristics of the input signals, even without the coefficients of the source signals and the transfer channels.In this dissertation, ICA and its application is introduced firstly. After reviewing existing algorithms of ICA, the research on blind separation of nonstationary signal is presented. As a new tool on nonstationary signals analysis, DWT based ICA is introduced. Simulations are made to measure the performance. The author brings a new scheme of digital watermark, in which ICA based embedding and ICA based detection is introduced as an application of ICA. While being embedded, an image is considered as a mixture of several independent feature causes, and Fastica is used as a powerful tool to get image feature . During detection, the ICA is used both on the original image and the watermark image respectively. Simulations show the scheme has good performance. It is robust to JPEG compression and other common image processing, including Gaussian noising, and so on. This algorithm can also be used in sound and video watermarking.