基于小波变换和广义有限自动机的视频压缩  

Video Coding Based on Wavelet Transform and GFA Modeling

作　　者：赵国锋[1] 康戈文[1] 孟继成[1] 

机构地区：[1]电子科技大学,四川成都610054

出　　处：《现代电子技术》2009年第20期63-65,共3页Modern Electronics Technique

摘　　要：为了改进视频编码的比特率,提出基于小波变换的位平面模型和广义有限自动机(GFA)的视频压缩。该算法是把一个连续画面中(GOP)的视频序列作为整体做小波变换,然后对小波系数的自适应量化及位平面模型。帧间,频带间,位平面间都类似成为相同位平面模型。这样可以更好地使用广义有限自动机(GFA)来压缩位平面。最后用熵编码方式编码GFA描述的节点信息为可分级的比量流,它是一个高效可分级视频编码方案。以前多数不切实际的多媒体通信应用将可能实现。In order to improve the video coding in bitrates,a video coding scheme based on the statistical bitplane modeling and the Generalized Finite Automata （GFA） representation of video sequences in wavelet domain is presented. In the proposed scheme,a video sequence Group of Pictures （GOP） is represented as an overall binary image by bitplane modeling the coeffi- cients of the video sequence within subbands using a generalized Gaussian quantizer,adaptive to the coefficient statistics of the video sequence. The inter-frame （motion） ,interband （frequency） and inter- bitplane （spectrum） similarities are optimally explored in the bitplane model,to form a compact GFA representation of the video sequence. Finally,all the transitions （edges） and the leaf nodes in the GFA representation are entropy encoded into a scalable bitstream. It is an effective scalable video coding scheme. Numerous multimedia communication applications, previously unpractical, should be envisaged with the low bitrates.

关 键 词：视频压缩 广义有限自动机 小波变换 位平面 

分 类 号：TN919.81[电子电信—通信与信息系统]