一种新的自适应的视频关键帧提取方法  

A novel adaptive video key frame extraction method

作　　者：王宇[1] 汪荣贵[1] 杨娟[1] 

机构地区：[1]合肥工业大学计算机与信息学院,安徽合肥230009

出　　处：《合肥工业大学学报（自然科学版）》2016年第11期1483-1487,1542,共6页Journal of Hefei University of Technology：Natural Science

基　　金：国家自然科学基金资助项目(61075032);安徽省自然科学基金资助项目(J2014AKZR0055)

摘　　要：针对目前关键帧提取存在的关键帧数目难以确定、对内容渐变的视频的处理效果欠佳和算法复杂等问题,文章提出一种自适应的视频关键帧提取方法,利用梯度方向直方图(histogram of oriented gradients,HOG)特征计算当前帧的代表性指标,初步选出候选关键帧,使用结合目标分割和颜色直方图的高级特征进行冗余度检查,最终确定关键帧集合。通过对大量包含各种场景的视频进行实验,结果表明使用该文算法提取的关键帧能更全面地表达视频的主要内容,尤其是处理内容渐变的视频时,效果更佳,并且能够根据视频的内容自适应地确定关键帧数目。In view of the problems in the existing methods of key frame extraction including the difficulty in determining the critical number of key frames, the unsatisfied treatment effect of the gradient of the content of video and the complex algorithm, an adaptive video key frame extraction method is proposed. Firstly, the value of representativeness of current frame is calculated by using histogram of oriented gradients（HOG） and the candidate key frame is selected. Then the redundancy check is made by using the target partition and color histogram features to determine the set of key frames. The ex- periments are carried out by means of a large number of videos containing various scenes, and the re- sults show that the key frames extracted by the proposed algorithm can describe the main content of the video more comprehensively, the effect is better especially when dealing with a gradient of video content, and the key frame number can be determined adaptively according to the content of the video.

关 键 词：关键帧 颜色直方图 梯度方向直方图 目标分割 自适应 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]