The Design And Achievement Of Video Watermark System Based On TMS320DM6437

With the the rapid development of multimedia technology and computer networks, digital media production and dissemination has become more convenient and faster, while piracy and copyright issues are becoming increasingly serious, multimedia information security has become a pressing need to be addressed. Because the digital products have widespread redundant and randomness data, the information such as digital, serial number, text, images and other signs are embedded in the products itself, so is digital watermarking technology. Now digital watermarking technology is more and more important, which will help a copyright protection, secret communications, and the authenticity of data files, and product identification and so on. Video watermarking technology is a hot and difficult point in current digital watermarking technology study. Potential to apply the video watermarking technology is very broad, including digital video-on-demand system, DVD copyright protection and satellite digital video transmission, etc.With the quick growth of digital image processing technology and DSP chinps, the digital signal processor, DSP has becomed to a basic device in communication, computer, and consumer electronics products. DSP chip plays a important role in core algorithm of video watermarking for DCT and IDCT transform, whose speed is crucial the entire system. How to choose a right video watermarking algorithm and design an efficient DSP platform software is a research difficult and focus nowaday's.This paper first takes an overview of the applications, characteristics, classification, and then describes the characteristics of DSP development. It also disscuss in details of the parameters of TMS320DM6437 DSP chips, set up a hardware system of video collection, processing and output. Algorithm of video watermarking is the most important thing of this paper. The author adopts a block-based adaptive algorithm of video watermarking. The core of this algorithm is the adaptive choice of the embedded area. For DCT transform has a great impact on system efficiency, the author uses a fast DCT algorithm, and transforms the complex floating-point DCT computation into smaller addition and shift computation, which greatly improved the effecency of DSP. Then by integrating paralled traits of fixed-point DSP with pipeling implementing ways, the time of watermarking algorithm will be much shorter than the time of a frame acquisition, so as to realize the realtime embedding of watermark. As to hardware, with a front-end high-speed FIFO buffers, video acquisition and output will be completed quickly through DSP-EDMA high-speed transmission.The most important feature of this paper is the realization an algorithm of realtime embedding of watermark in the fixed-point DSP video processing system. After running the program on the DSP platform, the experimental results show that the system can embed watermark to video in real time, and able to meet the requirements of general application.