Research On Digital Watermarking

As development and popularity of network and digital signal processing technique, multimedia information safety looks increasingly important. As a new branch of the information hiding technique, the digital watermarking is developed on basis of meeting requirement of protecting information copyright. It refers to embedding a mark (watermark) possessing special meaning into the multimedia products according to a given algorithm so as to realize protection of this products or authentication but not effecting the use quality of this product.This paper firstly presents the basic principle and signature of the digital watermarking as well as its researching status domestic and abroad, then gives out the basic frame of the digital watermarking, describes each composition in detail, and makes classification according to different signature of the digital watermarking, which provides theoretical base for raising, realization and detection of the watermarking algorithm. In embedding watermark, it presents DWT transform domain digital watermarking embedding technique in stress, because the DWT transform domain possesses advantageous of space domain and DCT transform domain, which can be used as data compression, processing and analyzing tools, and have good use in the digital watermarking. In addition, the paper has an independent chapter to present origin of chaos, mapping of the typical chaos and its usage in the digital watermarking. Chaos sequence can be utilized to provide a large number of noncoherent signal which is similar to random because of its sensitivity to the initial value. It can be used as the watermarking signal or is very useful for watermarking encrypting.In audio digital watermarking, this paper studies echo hiding watermarking algorithm in time domain in stress. It is different from the digital image watermarking. It is very difficulty to embed watermark in the digital audio signal. The main reason is that the human auditory system has higher sensitivity comparing with the visual system. The echo hiding audio watermarking algorithm makes use of backwards screening effect of human auditory system to audio signal in time domain, and it has advantageous of simple and easily generation.The main research contents in this paper are divided into following parts:1. Watermarking algorithm in wavelet transform domain research: this paper adopts two algorithms, one is using one dimensional chaos sequence as watermark signal to extract partial pixel block of the original image randomly to be used as a subimage to be embedded, use wavelet transform locally instead of globally so as to enhance secrecy. By comparing similarity of the extracted watermark and the original watermark, we can detect whether the watermark exists. The original image is not required to take part in detection, it belongs to blind watermark. The other is scrambling the grey watermark image so as to enhance secrecy by use Arnold transform, carry out wavelet decomposition of the carrier image in multiple layers so as to make the reconfigurable image with low frequency coefficient equal to original watermark image in size, then embed the watermark in the wavelet-decomposed low frequency coefficient. The original image is required to take part in detection, it belongs to nonblind watermark.2. Audio echo hiding watermarking algorithm research: using different echo delay represents different binary digital to add echo in segmented digital audio so as to achieve embedding watermark in echo. Detecting the information hided in the echo is to detect delay of the echo. This paper adopts the method of detecting peak value at delay sampling point of the autocorrelation complex cepstrum and power cepstrum curve to extract watermark.A large number of imperceptible and robustness experiments on the watermarking algorithm presented in this paper have been carried out. The simulation result shows that both of two image watermarking algorithms have perfect imperceptibility and strong robustness against the common image processing. The audio watermarking algorithm has perfect imperceptibility and antinoise detection performance.