基于曲波变换的渐变镜头转场检测算法  

Video shot detection algorithm based on curvelet transform

作　　者：宋志雄[1] 戴小文[2,3] 

机构地区：[1]西南交通大学交通运输学院,成都610031 [2]西南交通大学电气工程学院,成都610031 [3]厦门大学电子工程系,福建厦门361000

出　　处：《计算机应用》2009年第4期969-971,1021,共4页journal of Computer Applications

基　　金：信息产业部电子发展基金资助项目((2005)No.555)

摘　　要：曲波变换在处理"曲线奇异"时有明显的优势。给出了曲波变换的计算方法、公式和具体步骤。对dissolve渐变转场过程中的各相邻帧进行曲波变换。以曲波变换后高频系数的平方作为图像帧的特征量,计算相邻帧间此特征量的欧氏距离。用大量的高清码流对算法进行了实验。实验结果表明,此算法有较高的查全率和查准率。The advantage of curvelet transform in processing the ＂curve singularity＂ was introduced. Video shot gradual transition detection algorithm based on curvelet transform was presented. The method of calculation and formula was given out. The curvelet transform was used to dissolve image. The high frequency coefficients of curvelet transform acted as the characteristics of the image. The Euclidean distance between characteristics of the sequence images was proposed. The experimental results indicate that the algorithm has better performance than Y. H. HONG＇s scheme.

关 键 词：曲波变换 脊波变换 HAAR小波 检测 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]