基于h.264的時空域誤碼掩蓋技術(shù)的研究

1、華中科技大學碩士學位論文基于H.264的時空域誤碼掩蓋技術(shù)的研究姓名：程超申請學位級別：碩士專業(yè)：計算機軟件與理論指導教師：李勝利20090527華 中 科 技 大 學 碩 士 學 位 論 文 華 中 科 技 大 學 碩 士 學 位 論 文 IIAbstract In the process of real-time video transmission, data loss and transmission errors are al

2、ways inevitable. However, a small amount of errors may lead to incorrect decoding of a large number of codes in client. Furthermore, they may cause serious decline of video quality. To solve this problem, the error conce

3、alment technologies are used to recover the damaged information by using the temporal and spatial information redundancy. The method of error concealment based on client could not only recover the damaged block but also

4、reduce network overhead. When error occurs during the transmission, receiver uses the correct decoding information to improve image quality via reconstructed damaged block. Among them, spatial concealment recovers damage

5、d block using its neighboring blocks and temporal concealment makes use of temporal information redundancy to estimated motion vector (MV) of the damaged block and this MV is used to conceal the damaged block. But it can

6、not give satisfying results when using standard spatial and temporal error concealment algorithms. For the problems of the existing spatial error concealment algorithms, which could not recover the edge image well, an ad

7、aptive spatial error concealment algorithm (ASEC) is proposed. The directional information of the edges extracted from surrounding correctly received blocks are judged by using the directional entropy and edge strength,

8、and the damaged blocks are adaptively classified into uniform blocks and edge blocks. For the uniform blocks, the errors can be concealed by the bilinear interpolation algorithm; for the edge blocks, the errors can be co

9、ncealed by the directional interpolation algorithm based on region segmentation. Meanwhile, for the problems of the existing spatial error concealment algorithms, a multiple-reference temporal error concealment algorithm

10、 (MTEC) is presented to solve the problem of inaccurate of motion vector prediction. Firstly, the candidate motion vector set is determined; then the improved boundary matching algorithm and search algorithm are used to