| Video technology is an indispensable part of a multimedia technology. At the same time,the multimedia technical development has also pushed the video frequency technology to anew application altitude. However, due to the integration of wireless networks and multimediatechnology, originally we only need to meet the high video quality of multimedia technologyand starte to pay attention to the size of the amount of video data. In Wireless MultimediaSensor Networks, reduce the amount of data of the multimedia information has become themain purpose of the video compression coding.At present, H.264is the most widely used and the latest video coding standard, andcompared with other video coding standards, it has a higher encoding performance, while thishigh performance is established on the basis of its high complexity. Through the analysis ofthe H.264codec feature structure, we know that getting the predictive value is the mostcritical part of the H.264. Predicted value can be obtained through two paths, namely,intra-frame prediction and inter-frame prediction. These two processes are playing the keyrole in the entire code process, but the process is very complex, it occupies most of the timefor the entire coding process. Therefore, in order to improve the coding rate, the key is toimprove the efficiency of the prediction process.In the research process of H.264video coding algorithms, this paper mainly study onmotion estimation. Motion estimation as the dominant part of the forecasting process, itsefficiency directly determines the level of the entire coding process efficiency. In thearrangement of this chapter, it is first to analyze the framework of H.264codec, and conductsbasic research about the key technologies of the H.264encoding process. At the same time,because this article application background is the wireless multimedia sensor network, thisarticle selects the X264encoder, which has better practicality as well as is one of the threeopen source code in H.264. This paper is to analysis the X264four kinds of integer pixelmovement estimate algorithm in detail, and aims at UMHexagonS （asymmetrical cruciformmulti-level hexagon grid point search algorithm）algorithm to conduct the thorough research. Three optimization methods to make the coding efficiency enhanced by about20%isproposed, and still maintain the original coding performance and image quality.The ultimate purpose of video coding algorithm research is make it to the actual codingplatform.Therefore, in the last chapter, there is detailed study Linux embedded developmentstructures of the basic principles as well as the deve lopment environment, and transplants theX264encoder to the S3C2440platform realized video coding on ARM platforms. |
PDF Full Text Request