基于LDPC的分布式视频编码研究
基于LDPC的分布式视频编码研究(任务书,开题报告,外文翻译,论文12000字)
摘要
分布式视频编码（Distributed video coding, DVC）是一种新兴而又实用的视频编码，它基于上世纪七十年代Slepian-Wolf和Wyner-Ziv所提出的相关的信息编码论原理，具有编码简单，解码复杂，抗干扰强等特点。不同于传统视频编码（ISO/IEC MPEG和ITU-T H.26X），DVC可以大幅度降低编码端的复杂度，因而在小型或者低功耗便携摄像设备上具有广泛应用前景。虽然，关于DVC的研究和成果已经取得了不少进展，但是解码效果一直不是很理想，距离实际应用还有一定的距离，仍然有很大的提升空间。本文在部分前人研究的基础之上，提出了一种新的基于低密度奇偶校验码编码（Low Density Parity Check Code, LDPC），使用空间平滑和双向动作估计来提升边信息（Side Information, SI）质量的DVC方案——LSB方案（LDPC with Spatial Smoothing and Bidirectional Motion Estimation, LDPC-LSB）。本文主要工作包括：
1）简单介绍DVC的应用背景，当前国内外研究现状，同时介绍DVC的基本工作原理和理论基础。包括DISCOVER (DISTributed COding for Video sERvices, DISCOVER) 方案，基于像素域的Wyzner-Ziv编码方案等等，主要介绍这些系统的构成框架和各自的优势和不足。 [来源：http://www.doc163.com]
2）实现典型的基于经典Wyner-Ziv编解码方案的DVC方案的基本框架和工作流程，并仿真并分析这些方案的各自特点和不足。从这些方案的不足之处分析SI在整个分布式视频编码中的作用。
3）提出基于LDPC的LSB方案并详细介绍其所用技术：离散余弦变换，LDPC编码，空间平滑技术，双向动作估计算法，双向线性插值的工作原理，实现机制还有特点分析，并编码实现本文所提出的编码方案，并测试性能并与DISCOVER方案进行多方位比较。
本文主要设计了一个新颖的DVC编码方案（LSB方案），而测试结果也证明了它的提升：相对DISCOVER，在码率较低（200Kbps）时，LSB方案的峰值信噪比（Peak Signal to Noise Ratio，PSNR）有了1dB的提升；而在码率较高（350Kbps）时，PSNR则有了2.3dB的显著提升。从测试结果来看，LSB方案对比以前的方案所带来的提升还是比较可观的，是一个有效而又实用的编码方案。
关键词：分布式视频编码，DVC，LDPC，Wyner-Ziv，边信息
Abstract
DVC(Distributed video coding, DVC) is a novel and practical video coding, which is based on the theories about information coding and has some great features such as simple encoding, complex decoding and strong interference. Unlike traditional video coding (ISO/IEC MPEG and ITU-T H.26x), it decreases lots of complexities in the encoding side to make it work at ease on small, mobile and low-power-consumption devices. The researches and results about DVC have been obtained by many people, however, it still has many insufficient place. So in this paper, On the basis of previous studies, we propose a new scheme of DVC using Spatial Smoothing and Bidirectional Motion Estimation to improve the quality of SI(Side Information, SI), and we name it LSB Scheme. The main contents of this paper are listed as follows: [资料来源：http://doc163.com]
1) I will simply introduce the background of DVC, the progress of its researches in China and in the world, and some achieved results including DISCOVER (DISTributed COding for Video sERvices, DISCOVER) scheme and DVC based on pixel domain. And we will focus on their basic framwork, advantages and disadvantages.
2) And then, I will analyze the key of the improvement of DVC in detail, and explain why the key content of this paper is how to optimize the generation of SI and improve the quality of SI through the effect of SI in DVC.
3) Next, I will propose the LSB (LDPC with Spatial Smoothing and Bidirectional Motion Estimation, LSB) scheme based on LDPC(Low Density Parity Check Code, LDPC) and then introduce the operational principles, implements and features of these technologies used in detail including Discrete Cosine Transform, LDPC, Spatial Smoothing, Bidirectional Motion Estimation and Bilinear. In addition, I will write code to achieve them and make a test scheme for it to run performance tests, to compare with DISCOVER. [来源：http://Doc163.com]
In the paper, a better DVC scheme and its achievement are mainly proposed, and its test results show us the improvement. Compared to DISCOVER, while the bitrate is 200kbps,the improvement is just 1dB, but it go up to 2.3dB while the bitrate is 350kbps. The results of test show us that compared with old DVC scheme, LSB schemebrings more improvements in this paper is considerable, which is a better DVC solution for practical application.
Key Words: Distributed video coding, DVC, LDPC, Wyner-Ziv, side informatio
[资料来源：http://doc163.com]
目录
摘要 Ⅰ
Abstract Ⅱ
第1章绪论 1
1.1 研究背景及国内外研究现状 1
1.1.1 研究背景和意义 1
1.1.2 国内外研究现状 1
1.2 分布式视频编码基本原理 2
1.3 本文主要研究内容及结构安排 3
第2章分布式视频编码方案框架概述 5
2.1 分布式视频编码方案框架概述 5
2.2 分布式视频编码方案介绍 6
2.2.1DISCOVER方案 6
2.2.2 像素域编码方案 7
2.3 本章小结 8
第3章LSB方案框架以及算法详解 9
3.1DCT和LDPC介绍 9 [来源：http://www.doc163.com]
3.2 基于LDPC的码率控制方法 10
3.3 边信息相关算法 11
3.3.1 动作估计和运动补偿 12
3.3.2 双线性插值算法 13
3.3.3 空间平滑算法 13
3.4 本章小结 14
第4章LSB方案的实现和对比测试 15
4.1 编码环境简介 15
4.2 性能优化 15
4.3 测试方案和结果 16
4.4 本章小结 17
第5章总结与展望 18
5.1 工作总结 18
5.2 工作展望 18
参考文献 19
致谢 21