Research Of New Visible Watermarking

With the rapid development of Internet and multimedia technology, digital watermarkinghas become more and more widely applied in online multimedia authentication and copyrightprotection. Digital watermarking，which uses cryptography, wavelet analysis, human visualtheory, digital image processing, information theory and communication theory technology, isembedded in the multimedia contents. Unlike the invisible watermarking, the visiblewatermarking purposefully enables the visitors to see the embedded watermark, which moresuitable for copyright identification that has special requirement. This paper, which based on thedigital image of the visible watermarking, introduces two new visible watermarking algorithm.The unseen visible watermark, which is invisible under normal viewing pictures,can only showviewer in certain conditions,such adjusting the display viewing angle. The reversible visiblewatermarking, which bases on image fusion and pre-watermarking template,is working in a newimproved IDCT_DWT domain. The main works are as follows:1.This paper sums up several common methods about the visible watermarkprocessing,general watermarking framework and reversible watermarking framework.2. Simulation of the unseen visible watermarking proposed by Shang-Chih Chuang in LCDdisplay are realized with matlab.The embedded watermarking, which is invisible under normalviewing pictures,can show when viewer adjusted the display viewing angle. The algorithm isdiscussed further in the LED display. To have analyzed the features of LED monitor, thewatermarking,which can not be appeared because of the wide viewing angle features of LEDmonitor,can show by changing contrast of the picture in red color vector.3.This paper proposes a new improved IDCT_DWT algorithm. Simulation of the reversiblevisible watermarking is realized with matlab based on pre-watermarking template proposed theliterature [1]:High frequency templates, low frequency template and the correspondingembedding function. And compares the algorithm in DWT domain. Experimental data showthat,in the new improved IDCT_DWT domain,the peak signal to noise ratio is significantlyhigher than in DWT domain. At the same time we can get higher quality of restored image.