Research And Implement Of Image Quality Control And Multi-speed Services In Video Communication System Based On DM642

Along with the arriving of information age, the technologies of information processing, especially the video information processing, have been developed rapidly. The video communication system has been widely applied in video conference, video monitoring and other fields. When developing the video communication system, the hugest obstacle is the video acquisition, display and algorithm processing, because the video communication system need to be high-speed, steady and flexible, while handling the mass video data. Therefore, constructing the system for video acquisition, processing and display is becoming the key point of the actual application system design in WAN/LAN, and has drawn great attention.This dissertation proposes a scheme of video Multi-speed services and image quality control based on DM642 developing platform. This system acclimatize the image format and the frame rate to the parameter of the states of the net, and deal with the noise which shows up in the process of image zooming. So, the system may save bandwidth of the channel effectively and provide l services with different level of Qos.At present, there are lots of algorithms for image quality control aimed at aliasing noise, but not enough for real time video communication system based on DSPs. Because the limited operaten ability of DSPs, most of the system select a algorithm of anti-aliasing with lesser calculate burthen. The method of anti-aliasing concludes two kinds: space-dimension and frequency-dimension. We often translate the image from space-dimension to frequency dimension with Fourier Transform, Wavelet Transform and so on. Fourier Transform is a classic analysis tool in image processing skills. Wavelet Transform is a rising method of image processing. Because of the special space-frequency characteristic of Wavelet Transform, we can analyze and describe the object in different measure, and Wavelet Transform has becoming to an excellent tool of anti-noise.Aim at anti-aliasing in the spatial dimension; we usually make use of theSpatial Lowpass Filter. Using it's different form to convolute with image Matrix.Server of this system is built on TI DM642 video processing platform. TI DM642 is a kind of DSP chip designed for multimedia applications. Its core clock is up to 600MHz, and it has 8 parallel operaten units, its processing ability is up to 4800MIPS. For multimedia applications, DM642 video processing board integrates 3 configurable video ports, McASP for audio applications, 10/100Mb/s Ethernet MAC, and so on.With regard to the hardware, according to the hardware characteristic of DM642 platform. The program is composed and the library files (CSL and BSL) are used to configure the SAA7115 chip through I2C bus. The system can provide two-way video acquisition and NTSC/PAL color image, and implements the board extending camera's acquisition of D1/CIF format. Besides, it can resample the image according to the feedback information of the network condition, so the CIF/QCIF/SQCIF/user-defined image can be obtained to reduce the pixel number in one frame. When displaying, it is also necessary to configure SAA7105 chip, thus the system can provide both the demanding size image of current network condition and the CIF size image after interpolation, then display these images in four divisional areas of screen through the board extending TV.In the process of image re-sampling, the aliasing noise appears unavoidably. For the server side, considering the operaten ability and the quest of video communication system, this system selected some low-rate video format (D1, CIF, QCIF, SQCIF) which clamed by CCITT. Aiming at those video format, this dissertation estimates the calculate capacity of filtering, and then select two-dimension Fourier Transform and fast median filter for the process of anti-aliasing. Considering the separability character of Fourier Transform, this paper use one-dimension Fourier Transform twice to complete the two-dimension Fourier Transform of the image Matrix. Aiming at the video format in this system, we use Zero-Complement method to implement the two-dimension Fourier Transform to the image Matrix in DM642 platform. In order to satisfy the quest of real-time video communication system, this dissertation improve the algorithm of two-dimension Fourier Transform ,take full advantage of DM642DSPs, selectproper way of translate and edit and improve the speed of this system obviously.For the client side, in order to satisfy the sight of customer, this dissertation select double-liner-interpolation algorithm to control the format of video image. Considering the character of the video format in this system, this dissertation brings forward the fast algorithm in DM642 platform and improves the speed of operaten. Double-liner-interpolation efficienctly conquers the disfigerment of ordinary interpolation algorithm and improve the quality of image greatly.For QoS realization of video communications, control of video transmission is very important. For video transmission, primary characteristic of channel in application layer is available bandwidth and packet loss. In the case of packet loss and variable bandwidth, QoS in application layer can promote the quality of video in furthermost. Rate control attempts to minimize network congestion and the amount of packet loss by matching the rate of the video stream to the available network bandwidth. This paper mainly discusses realization of source-based rate control. To reduce frequency of video sampling can reduce output video data rate and frame rate. In this way, adjusted output data rate and frame rate can be suitable for different network bandwidth. Moreover, to reduce sending frame rate also can reduce sending rate. Video communication system can monitor QoS using sender report and receiver report of RTCP. If RTP packet loss rate is higher than some threshold, video data rate should be reduced. If interval of receiver report longer than some threshold, loss rate section of receiver report should be checked. If the section shows network congestion, data rate and frame rate should be reduced.The technology of image quality control and Multi-speed services based on DM642 video communication system is a new research field, the system proposed by this dissertation has not fully developed the huge potential of DM642 yet. There are still some areas that this dissertation didn't touch. Moreover, the current technology may also need to be improved and advanced. The technology of image quality control and Multi-speed services based on DM642 video communication system has a promising research future, and it is also a vigorous and exploitable field. This dissertation makes an outlook for the technologies of image quality control and Multi-speed services based on DM642 video communication system inthe end.