基于脊波变换的渐变镜头转场检测算法  

An Algorithm of Video Shot Detection Based on Ridgelet Transform

作　　者：戴小文[1,2] 宋志雄[3] 钟桂英[4] 

机构地区：[1]西南交通大学电气工程学院 [2]厦门大学电子工程系,福建厦门361000 [3]西南交通大学交通运输学院 [4]西南交通大学建校指挥部,成都610031

出　　处：《光子学报》2009年第5期1297-1300,共4页Acta Photonica Sinica

基　　金：国家自然科学基金(50407009)资助

摘　　要：分析了脊波变换在处理含"线奇异"信号上的优势.提出用脊波变换后的各高频系数和作为图像的特征量.渐变镜头脊波变换的实现分两步:首先,对图像帧进行一定方向参量的Radon变换;其次,对Radon变换得到各数组进行Haar小波二级分解,采用分解后的高频系数和描述图像.在检测窗内,计算连续帧间脊波变换系数和之间的欧式距离判别渐变转场的边界.The advantages of ridgelet transform in processing the “Linear Singularity” were analyzed. A new video shot gradual transition detection algorithm based on ridgelet transform was present. The sum of ridgelet transform high frequency coefficients acts as the characteristics of the image. Radon transform of image was used,and Haar wavelet transform was given out. The Euclidean distance between the sequence images was given out. Ridgelet transform coefficients were used to detect the gradual shot-boundary. The experiment results indicat that the new algorithm givens better performance.

关 键 词：渐变镜头 脊波变换 RADON变换 HAAR小波 检测 

分 类 号：TP751[自动化与计算机技术—检测技术与自动化装置]