N/2帧回退与十字切片相结合的镜头边界检测算法  

An Algorithm of Shot Boundary Detection of Backing N/2

作　　者：吴开兴[1] 沈志佳[1] 

机构地区：[1]河北工程大学信息与电气工程学院,邯郸056038

出　　处：《科学技术与工程》2014年第4期250-254,共5页Science Technology and Engineering

摘　　要：基于内容的视频检索是当前计算机领域研究的热点之一。在基于内容的视频检索中,镜头边界检测技术是核心关键的技术之一。文中针对镜头边界检测,在传统算法的基础上提出了一种N/2帧回退与十字切片相结合的镜头边界检测新算法。对该算法利用MATLAB仿真并通过与传统算法对比,结果表明,算法复杂度大大降低,镜头边界检测的时间缩短了一倍以上,并且在一定程度上提高了视频镜头边界检测的准确率。Content-based video retrieval （CBVR） is one of the research hotspot in computer science. The shot boundary detection is one of the key technologies in CBVR. Based on the traditional algorithm, an algorithm of shot boundary detection of backing N/2 combining with cross chop is proposed. Simulating the new arithmetic through the MATLAB, and comparing it with the traditional arithmetic, it shows that the new arithmetic greatly reduces the algorithm complexity, and the time for shot boundary detection reduces more than half. Except that this algorithm improves the accuracy of shot boundary detection somewhere.

关 键 词：视频检索 镜头边界检测 N 2帧回退 十字切片 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]