Research On Multimedia Self-Adaptating Coding Transmittion Technology

In image and video coding, multiple descriptions coding algorithm has good resistance to jamming and error, for lower packet loss fewer data errors and other problems, which caused by the unreliability of channel in wireless communication network and Internet. This paper reviews the state of the art of the multiple descriptions coding, sand gives three novel multiple descriptions image coding algorithms and their simulation experiments.The first algorithm is based on low frequency prediction. First, the wavelet coefficients of original image are grouped in different ways according to the low and high frequency; then. because of the low frequency coefficient is critical to the restored image quality, preprocessing and post-processing based on the prediction are used to achieve redundancy insertion between the descriptions; finally, error concealment algorithm is used to improve the robustness of the transmission.The second algorithm is a two-stage multiple descriptions coding based on the wavelet domain. First, the wavelet coefficients of the original image are coded with SPIHT coding algorithm to be formed the first part of the descriptions, which be copied and synchronize delivered to two channels; second, the coefficients are decoded after the SPIHT coding, and doing difference with the original wavelet coefficients; third, the difference signal are sampled by the quadtree method, and coded with SPIHT algorithm in the bit rate, which formed the second part of the descriptions; finally, the two descriptions are transmitted in two channels respectively.The third algorithm is an equilibrium multiple descriptions coding based on DCT. First, sixteen low frequency coefficients are extracted from8×8DCT coefficients block, the four low-frequency coefficients of which are assigned to the two descriptions respectively, the remaining twelve coefficients alternately assigned to the two descriptions by zigzag sequence scanned method, which formed the base layer BL1and BL2; then, sixteen high frequency coefficients in the horizontal and vertical directions are directly assigned to both descriptions, which formed the enhanced layer EL1and EL2; finally, BL1, BL2, EL1and EL2are combined to form four descriptors, transmitted through different channels.Finally, the proposed equilibrium multiple descriptions image coding based on DCT is applied in the development of real-time monitoring module of the online purchase supermarket system, according to the comparison results of the three above multiple description coding schemes,. The results show that the quality of the decoding image improves with the increasing number of receive descriptions, especially in color. It is sufficient meeting the demand of the server-side monitoring. This proves the effectiveness of the proposed algorithms, which provide a viable solution for the realization of adaptive transmission of image and video coding.