基于小波的视频空间可分级编码研究

Research on Spatial Scalable Video Coding Based on Wavelet

1.引言 在许多实际应用中,由于用户要求不同、终端能力不同、异构网络的不同支路所能提供的QoS不同、或网络传输条件的变化(比如噪声、拥塞等)等原因,需要提供不同质量的视频信号,比如视频会议、视频点播、多媒体数据库浏览等等,解决此类问题的最好方法是用单个编码器产生分层次的压缩码流,对不同层次的码流解码可以获得不同的图像质量,这种编码机制通常被称为可分级编码(Scalable Coding).

New applications in a multiresolution environment arise the need of video scalable coding. The video scalable coding based on wavelet transformation has received much attention in recent years. Based on a thorough examination of current research works on motion estimation and prediction, this paper gives a kind of video spatial scalable coding scheme based on a kind of indirect approach, which uses the motion prediction on lowpass image at the next finer resolution to instead of predicting on the highpass subbands at a given resolution, to avoid the motion tracking errors in high frequency bands. Simulation results show it is effective.