节点文献
基于颜色和灰度熵的视频镜头检测方法
Video Shot Boundary Detection Based on Color and Gray-scale Entropy
【摘要】 视频镜头分割是视频检索和视频分类等应用的关键步骤,因此提出一种基于颜色直方图和灰度熵的镜头检测算法。该方法以HSV颜色直方图和图像的灰度熵为特征,计算视频相邻两图像帧的相似度,从而判断出是否发生了镜头变换。实验证明,该方法能准确检测出视频的突变镜头,对渐变镜头也有理想的检测效果。
【Abstract】 Video shot segmentation is a key step in the application of video retrieval and video classification.This paper presents an effective shot detection based on color histogram and gray-scale entropy.The method is characterized by a non-uniform quantized HSV color histogram and gray-scale entropy.Then it calculates the similarity of two adjacent frames of the video to determine whether the video transforms to another shot.The experiments results indicate the method can accurately detect abrupt cuts,and also has ideal effect on detecting gradual cuts.
【关键词】 HSV颜色直方图;
灰度熵;
相似度;
镜头检测;
【Key words】 HSV color histogram; gray-scale entropy; similarity; shot detection;
【Key words】 HSV color histogram; gray-scale entropy; similarity; shot detection;
【基金】 北京邮电大学青年科研创新计划专项课题(G470289)
- 【文献出处】 电视技术 ,Video Engineering , 编辑部邮箱 ,2012年21期
- 【分类号】TP391.41
- 【被引频次】1
- 【下载频次】71