节点文献

H.264可伸缩视频编码码率控制算法研究及其应用

H.264 Scalable Video Coding Rate Control Algorithm Research and Application

【作者】 杨金

【导师】 孙世新;

【作者基本信息】 电子科技大学 , 计算机软件与理论, 2011, 博士

【摘要】 随着视频压缩技术的发展,最新的H.264可伸缩视频编码技术(H.264/SVC)提供了在时间、空间和信噪比不同级别上的可伸缩性。H264/SVC使用了分层的结构,每个SVC视频比特流由一个基本层和一个或多个增强层组成。每个增强层要么提高视频(空间或时间)分辨率要么提升视频序列的质量,所以SVC数据流的比特流可以适应于不同的传输环境、不同的视频应用,以及不同的网络要求。与此同时,随着多媒体数据的使用显著增长,作为在无线通信中最重要的突破之一,多输入多输出(MIMO)技术通过在发射和接收端设置多个天线来增大无线信道的吞吐量,可以提供更高的数据吞吐量和更低的网络延迟。由于分层的H.264/SVC视频流从本质上适合在MIMO信道上并行传输,SVC与MIMO技术的自然结合能显著提高无线视频应用的整体性能。当压缩视频用于无线网络传输时,过高的码率输出会导致网络拥塞和视频数据包的丢失,同时过度减少输出比特率则可能导致不必要的质量退化和可用网络带宽的浪费。码率控制算法(RC)必须根据可用的网络带宽控制视频编码器输出的比特率,同时获得最佳的视觉质量并保持缓冲区充盈度在一个可接受的范围之内。因此,码率控制算法在视频压缩和传输中都起着重要的作用。为了实现H.264/SVC视频编码技术与MIMO无线系统的有效结合,本文全面系统地研究了H.264/SVC可伸缩视频编码码率控制算法,以及H.264/SVC-MIMO系统的联合控制算法,设计和实现了一个高效的实时无线视频压缩传输系统。本文的主要内容及创新点如下:(1)分析了H.264/SVC可伸缩视频技术的基本原理以及现有的各种码率控制算法的优缺点。(2)首先提出了H.264可伸缩视频编码的初始帧QP计算方法。该方法相对于现有的方法提供了更为精确的初始QP值估计。(3)进一步提出了H.264可伸缩视频编码的基于增量的码率控制算法,相对于H.264/SVC参考软件JSVM中使用的码率控制算法,本文中的算法不仅实现了基本层的码率控制,而且实现了时间、空间和质量增强层的码率控制,并提供了更为精确的输出编码比特率和更好的编码质量。(4)在基于增量的H.264/SVC码率控制算法的基础上,对Inter帧码率控制进行了进一步研究,提出了自适应的Inter帧层间码率控制算法。降低了视频中的画面突然剧烈改变对码率控制算法的影响,提高了控制的精确性。(5)为了实现H.264/SVC与MIMO无线通信系统的结合,提出了H.264可伸缩视频流的抽取和重组算法,该算法可以将不同的可伸缩层分别抽取出来成为独立的数据流,通过MIMO无线传输系统的不同子信道上进行传输,并在接受端重组为可解码的数据流。(6)最终将H.264可伸缩视频编码的码率控制算法与MIMO无线传输系统相结合,通过将MIMO信道状态的预测及时反馈给编码器进行源-信道联合码率控制,实现了H.264/SVC数据流在MIMO无线系统上高效稳定的传输。

【Abstract】 As the latest advance of video compression technology, the emerging H.264 Scalable Video Coding (H.264/SVC), an extension of H.264/AVC standard, has improved coding efficiency and provided a higher degree of scalability in temporal, spatial and SNR, when compared with previous scalable video coding schemes. The SVC design enables the creation of a video bit stream that is structured in layers, consisting of one base layer and one or more enhancement layers. Each enhancement layer either improves the resolution (spatially or temporally) or the quality of the video sequence. Meanwhile, the demand for high-quality mobile wireless communication services is increasing at an explosive rate primarily due to significant growth of multimedia data usage. As one of the most significant breakthroughs in wireless communications, Multiple Input Multiple Output (MIMO) technology, which puts multiple antennas at both the transmitter and the receiver to multiply throughput of a radio link, can provide higher level data throughput with extremely low latency. With a layer structure, SVC can gracefully make bitrate, format and power adaptation in lossy transmission environment to accommodate various video applications as well as diverse network requirements. As multi-layered H.264/SVC bitstreams are inherently suitable for parallel transmission over MIMO channels, it is undoubted that the natural integration of SVC with MIMO technique can significantly enhance the overall performance of wireless video applications.When compressing video streams for network transmission, over-rating the output of a video encoder may lead to network congestion and video data loss; while over-reduction of the output bit rate may result in unnecessary quality degradation and waste of available network bandwidth. Rate control (RC) must then be adopted to regulate the output bit rate of a video encoder to obtain optimum visual quality within the available network bandwidth and to maintain buffer fullness within a specified tolerance range. Therefore, RC plays a crucial role in video compression and communication.To integrate H.264/SVC with MIMO wireless system, this paper researchs the H264/SVC rate control algorithms and Joint H.264/SVC-MIMO rate control schema step by step, finally, we provide an efficient realtime wireless video compression and transmission system.The core content and novelties of this paper are as the follows:(1) This paper analyzes the principles and coding schemes of the SVC technologies, and existing Rate Control schemas.(2) This paper proposes a H.264/SVC initial QP calculation method for Intra frame, which provides more accurate initial QP estimation than existing methods.(3) This paper proposes an incremental H.264/SVC rate control algorithm. Comparing with the rate control algorithm using in H.264/SVC reference software JVSM, our rate control algorithm is not only able to handle basic layer rate control, but also all enhancement layers, while provide more accurate output bitrate and better quality.(4) Based on the incremental H.264/SVC rate control, this paper proposes an adaptive inter-layer rate control algorithm for Inter frames, which reduces the risk for abrupt change in video sequence and provides more accurate output.(5) To integrate the H.264/SVC rate control with MIMO wireless system, this paper proposes a extracting and re-assamble algorithm for H.264/SVC bitstream, which extracts different separate layer of H.264/SVC bitstream into sub-streams transferring on different MIMO channel, and re-assamble the sub-bitstreams into decodable SVC bitstream on the receiver side.(6) In the end, integrating the H.264/SVC rate control algorithm with MIMO wireless system, this paper proposes a Joint Source-Channel rate control algorithm, implements an efficient and stable wireless video compression and transmission system.

节点文献中: 

本文链接的文献网络图示:

本文的引文网络