| There are many new technology in 21 century in which the technology obtain high-speed development. And the information that based on computer, net and electron technology obtain advance rapidly development. Video frequency technology has a rapid development from black-white to color and to today high definition digit frequency. The video frequency technology is developing to digit, high definition and three-dimensional. Multi-angle video frequency is an important development in future video frequency. 2-D color movement image has better vision effect in many management today, but the need is higher and higher with the developing of the standand of living. The image and video is popular to people for its third dimension. People haven't satisfied with watching the television passively, they demand vision media which is flexible.The third dimension of the video image makes the storage and transmission of video data hard, and good bandwidth is highly needed in the timing transmission of the solid video. The image communication of high-distinguishing and multi-view point-to-point maybe is the main reason for the future net congestion. The huge data makes the storage of memory, channel transmission rate of trunk line and the speed of computer work hard, which is the bottle-neck problem in the implement of multi-video application. We have two solutions to solve the problem. Firstly, we extend the storage of the memory, increase the channel transmission rate of trunk line and improve the speed of computer, which is hardly to realize for the performance of hardware is limited. Secondly, we reduce the quantity of data, store and transmit the data in the form of compress in order to relax the technology pressure radically which the huge data brings. We get rid of the redundant information and keep down the independent information part so as to compress the data, in which the best solution is compressing and coding for the video information.Therefore, it will be an important research area for the key of efficiently data compression encoding. Although the two-dimensional picture compression has already made huge progress in the theory and the practice, it is not mature in the three-dimensional effect and the multi- views- transmission aspect, how to encode the multi- views video image and reduce the data quantity of multi- views video will be the key of multi- views video towards practical. Now, the most research of multi- views video is stereo video , the traditional algorithms of multi- views video encoding is based on block and based on content. The advantage of algorithm based on block is high steadiness ,easy hardware realization and simple algorithm, but it will appear the blocking effect, and weak subjective effect of reconstructs picture under low bit rate. The content-based encoding algorithm has widespread interest in recent years, and is thought as the substitution algorithm for the block-based algorithms. The advantage of content-based encoding algorithm is the low encoding mistake, good visual effect, the constitutive description of scene and adaptation of new domain, but it is also in the initial stage, the technology of it is not mature. Based on this, a new compression encoding algorithm based on five dimensional matrix multi-views color video is proposed in this paper.The compression encoding algorithm based on 5DM-DCT multi-views color video in this paper is used to realize the compression encoding of stereo video. Because the stereo video consist of two channels, there are two important redundant information in stereo video -motion vector and parallax vector. Former is used to describe the relevance in time domain of continue frames in the single channel, But latter is used to describes the relevance in space domain between different channel projection images in the same time and scene. Therefore, in order to fully considers of the relevance in time domain between frames, the color relevance of image in frame and the space relevance between two channels, a new compression encoding algorithm based on five dimensional matrix multi-views color video is proposed in this paper- Compression Encoding Algorithm Based on 5DM-DCT Multi-Views Color Video. It uses a new mathematical model- five dimensional matrix to express the stereo video, which is expressed in a unification mathematical model. DCT of five dimensional matrix is used in this new transformation method in order to uniform the color video in different channels and the process of many frames in a channel, then transform them into algorithm processing of five dimensional matrix. This algorithm not only comprehensively removes the relevance between various gradation images of color video in one channel, three frames in color space and the continuous video frames, but also removes the relevance between different channels in space. Thus it realizes the high time of compression under the high signal noise ratio condition. In terms of Stereo Video, the concrete steps of Multi-View Color Video Coding methods are as follows :1. Three-dimensional video signal has been expressed with a five-dimensional matrix, and the five-dimensional matrix of Color Stereo Video has been divided into "five-dimensional block", namely the five-dimensional sub-matrix, the size of the five-dimensional sub-matrix can be set on according to different application;2. The five-dimensional sub-matrix that has been divided well is transformed on according to the discrete cosine transform method of five-dimensional matrix and the five-dimensional sub-matrix of the transform coefficient can be gotten;3. After 5DM-DCT of the five-dimensional sub-matrix, the sub-matrix of the transform coefficient has been gotten which has already cut some redundancy. And then the sub-matrix of the transform coefficient has been quantified, in order to remove the redundancy that has no effect on the image; After the quantification of the transform coefficient of 5DM-DCT, the coefficient that has been already quantified will be encoded and then the code will be encoded with entropy on the basis of Statistical Properties-Huffman encoding. The stream of codes that have been encoded has been transferred through channel or has been memorized in the memory media;The process of decoding is in contrast with the encoding.Secondly, this paper has adopted the methods of descending compression of the multiple view color image, and the simulative experiment has been done; because there are no unified international standards on the field of multiple view video compression encode, the 4DM-DCT has been applied in the stereo video compression encoding in this paper in order to measure the compression capability of the multiple view video compression encoding on the basis of 5DM-DCT,this paper has given the methods of stereo video compression encoding, and the situation of the video image being uniform or complex is considered in the simulative experiment, through the comparison of the experiment result, the compression encoding method of the multiple view color video stream on the basis of 5DM-DCT has better compression capability.Multiple view video is a fashion research issue, the research contents are very wide, there are many difficulties in the theory and practice that have been settled in the aspect of the multiple view video application, much research work will be done. |
PDF Full Text Request