摘要
在传输中尽可能保持视频质量的前提下 ,维持视频流码率的恒定 ,是码率控制的核心内容 .一般 ,传统的方法需要在做量化之前计算数据的 R- D特征值 ,以确定量化策略 ,计算复杂度高 .这对于要求实时编码的场合显然是不合适的 .本文首先建立码率失真模型 ,通过在模型上求失真最小化 ,得到量化参数与目标码率之间的函数关系 ,然后针对低比特传输简化 ,形成适用于系统的快速方法 .最后 ,本文给出在 PC上仿真该算法的结果 ,证明这种方法比 TMN8模型算法有更严格的码率控制 ,同时可以用于实时应用 .
Its the most important to keep video transmission bitrate constant while not degrading video quality. Traditional approaches are so complex because they require the R D characters of input data to be measured before making quantization assignment decisions. Its computational complexity is not suitable in real time application. This paper proposes one solution based on R _D model, and gets the correlativity between quantization parameter and aimed bitrate by minimizing the distortion. After predigesting, this becomes one fast bitrate controlling method. At last, we give the test result emulating this method on PC. It obviously seems that the proposed method is more veracious on rate controlling than that of TMN8 and can be used in real time application.
出处
《小型微型计算机系统》
CSCD
北大核心
2001年第6期687-689,共3页
Journal of Chinese Computer Systems
基金
国防科技预研任务"评估多媒体数据存取与交换方法的研究 15 .8.4"和"2 11工程"重点学科建设项目"数据存储技术和系统结构"的资助
