基于双重检测的视频镜头分割方法  

Video shot segmentation method based on dual-detection

作　　者：杨瑞琴 吕进来[1] YANG Rui-qin;LYU Jin-lai(College of Computer Science and Technology, Taiyuan University of Technology, Taiyuan 030024, Chin)

机构地区：[1]太原理工大学计算机科学与技术学院,山西太原030024

出　　处：《计算机工程与设计》2018年第5期1393-1398,共6页Computer Engineering and Design

基　　金：山西省2013年科技攻关基金项目(20130321007-02)

摘　　要：为准确检测出视频的镜头边界,有效区分突变和渐变镜头,提出一种基于双重检测的镜头分割方法。该方法分初检和复检两个过程,初检是对视频帧进行目标与背景的分离,对目标利用Hu不变矩提取特征并进行检测;复检采用三维颜色直方图对初检结果进一步检测,得到镜头边界;根据突变和渐变镜头的过渡时间不同的特点,提出利用滑动窗口对镜头的过渡时间进行检测。实验结果表明,该方法的查全率和查准率均超过了80%,与其它方法的检测结果相比具有较高的准确性。To accurately detect the video’s shot boundary and distinguish abrupt and gradual of shots effectively,a video shot segmentation method was proposed based on dual-detection.It consisted of two processes including pre-detection and re-detection.At pre-detection stage,the video frames were divided into the target and background.Features of target were extracted using Hu moment invariant and the dissimilarity was detected.At re-detection stage,the three-dimensional color histogram was used to further detect the results of first stage and the shot boundary was got.According to the different transition time of the abrupt and gradual of shots,the sliding window was used to detect the transition time of shots.Experimental results show that the recall and precision of method are more than 80% .Compared with other methods,the proposed method has high accuracy.

关 键 词：双重检测 HU不变矩 三维颜色直方图 滑动窗口 突变 渐变 

分 类 号：TP37[自动化与计算机技术—计算机系统结构]