基于感兴趣区域的小波变换可分级视频编码研究  

Research on Wavelet Transform Scalable Video Coding Based on ROI

作　　者：吕宝成[1] 张贵仓[1] 朱贤坤[1] 

机构地区：[1]西北师范大学数学与信息科学学院,甘肃兰州730070

出　　处：《计算机技术与发展》2008年第12期81-83,共3页Computer Technology and Development

基　　金：国家教育部重点科学技术项目(204143)

摘　　要：随着Internet和无线通信的飞速发展,人们在网络上实时获取视频数据已经成为可能。由于传输网络和接收终端的多样性,所以需要视频流能适应多种不同传输、解码和显示的要求,由此产生了可分级视频编码。文中在可分级视频编码的基础上对其进行改进,把感兴趣区域的检测和小波变换应用到可分级编码中,该方法中对视频流进行感兴趣区域检测,利用小波变换对增强层中的感兴趣区域进行编码,由于小波变换具有良好的空间方向选择性,与人眼的视觉特性十分吻合,从而得到很好的效果。With the development of Intemet and wireless communication,it is easier for people to obtain real - time video over networks. The diversity of network and receiving terminals, so that it requires video stream should have the ability to adapt itself to various transmission, decoding and display requirements. So results in scalable video coding. The improvement that the extraction of the regions of interest and the wavelet transform are applied to the scalable video coding is done based on scalable video coding in this paper. Extracting the regions of interest from video stream in this method. Using wavelet transform to code the region of interest of the enhance layer. As wavelet transform with good spatial orientation selectivity, and the human eye in line with the visual characteristics, the experiment proves that the method is efficient.

关 键 词：感兴趣区域 可分级视频编码 人类视觉系统 小波变换 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]