一种基于时空信息的多目标检测新算法  

Novel algorithm for multi-target detection based on spatial-temporal information

作　　者：赵钦君[1] 赵东标[2] 陆永华[2] 

机构地区：[1]南京航空航天大学自动化学院,南京210016 [2]南京航空航天大学机电学院,南京210016

出　　处：《仪器仪表学报》2011年第4期877-882,共6页Chinese Journal of Scientific Instrument

基　　金：国家自然科学基金(51005122);航空科学基金(2008ZE52049)资助项目

摘　　要：运动目标检测是计算机视觉中进行信息提取的关键技术之一。为了克服传统帧间差分法检测出的目标不完整的缺点,提出了一种基于时空信息的视频图像中多目标检测新算法。该算法将帧间差分得到的图像序列的时间信息和帧间边缘差分得到的图像空间信息结合起来检测图像中的运动目标,同时利用形态学闭运算和近区域合并方法进一步强化图像中的空间信息,利用小区域去除的方法去除图像中的噪声影响。最后通过3组实验对其有效性进行了验证,实验结果证明,其检测效果与传统算法相比有较大改善。Moving target detection is one of the key technologies for information extraction in computer vision.In order to overcome the defect that conventional temporal difference method can not detect complete target,a novel algorithm for multi-target detection in video images based on spatial-temporal information is proposed.The algorithm combines the temporal information obtained using temporal difference with the spatial information obtained using frame edge difference to detect moving objects in images.What＇s more,the spatial information in the image is strengthened using morphological close operation and close region merging,and the noise influence to the images is eliminated using small region removing method.At last,the effectiveness of the algorithm is verified with three groups of experiments.Experimental results prove that the detection effects of the proposed algorithm are greatly improved compared with conventional algorithms.

关 键 词：时间差分法 时空信息 多目标检测 形态学 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]