Research Of Low Bit Rate Video Compression And Realization Project

With the development of image information technique, the video image compression code technique is mainly applied in the low code rate at present. Such the technique as videophone, videoconference, long-distant monitoring, can bring about enormous image information, compressed data, and transmission with high quality. The goal of this compression code technique is making the details of compressed image be as true as the ones of described scenery, and making image life sive and bandwidth the least during storage and transmission.Cancelling frame redundant through motion estimation and compensation is one of the key technique of the video image compression code. Since there are a lot of methods to compress the frames, different methods should be adopted correspondingly to different fields. The request of data compression with lowest coding rate is reducing the code rate using the relevant characteristic of frame while keeping high fidelity of image. Our study finds out that the compensatory system of relevant frame throwing frame can enhance the rate of compression and improve the quality of image transmitted by the lowest code rate. (1) SPECK based on wavelet transform compressed inside the frame has been led into. The advantages of upgrade algorithm on the aspect of being structured easily, analyzed from multiple angles and embedded code which is little counting, strong correcting help to overcome "the diamond effect " being resulted by DCT. (2) The thesis puts forward the algorithms of closing time redundant between neighboring frame in rhomboid research and the model of judging the thrown frame by analyzing the error, according to the visual sense, the study of relevant frame, and the comparing analysis through motion estimation and. The passage also discusses different strategies and compensatory methods in thrown frame, and tests this algorithm in the applied field of low frame ratio. (3) This thesis designs the scheme of image compression code based on Nios adopting technique of SOPC. This scheme, which is consistent of collecting images, dealing with image compressed code, and exporting from video has characteristics of being customized, reconfigurable, design flexibility and programmable.The video image coding algorithms by software adopts the method of uniting the code inside and between the frames,designs the sample frame transported by itself and controlled by bit rate. It also avoids image distortion for accumulating the errors, fitdifferent require of frame ratio and bandwidth, and obtains compression rate at any time. Comparing with the simulation in accordance with H.263 standard, the scheme is verified with high compression efficiency, fast running speed without giving no attention to the balance of compression rate and image quality, which is the research with practical value in the low bit rate video image compression field.