基于OpenCV的运动目标识别跟踪  

Moving target recognition and tracking based on OpenCV

作　　者：石泽琼 毛俊强 李双伟 徐丽蕊 董言治[1] SHI Ze-qiong;MAO Jun-qiang;LI Shuang-wei;XU Li-rui;DONG Yan-zhi(School of photoelectric information science and technology,Yantai University,Yantai 264000,China;Yantai Haibo Electrical Equipment Co.,Ltd,Yantai 264000,China;Yantai Unicair Communication Tec Co.,Ltd,Yantai 264000,China)

机构地区：[1]烟台大学光电信息科学技术学院,山东烟台264000 [2]烟台海博电气设备有限公司,山东烟台264000 [3]烟台友开通信技术有限公司,山东烟台264000

出　　处：《中国集成电路》2019年第4期20-25,共6页China lntegrated Circuit

摘　　要：实时目标检测与跟踪系统是近年来计算机视觉与模式识别领域的研究热点之一。本文主要是在Visual Studio2012平台下,使用计算机视觉库OpenCV对运动图像进行形态学图像预处理,以及调用CamShift算法对实时运动目标进行识别跟踪。该算法首先计算目标是HSV空间下的HUE分量直方图,通过直方图反向投影得到目标像素的概率分布,再调用CV库中的CamShift算法,自动跟踪并调整目标窗口的中心位置与大小。实验结果表明,该算法性能良好,能够有效地进行实时目标追踪。real-time object detection and tracking system is one of the hot spots in computer vision and pattern recognition in recent years.This paper mainly USES OpenCV,a computer vision library,to preprocess morphological images of motion images under the Visual Studio2012 platform,and calls CamShift algorithm to identify and track real-time moving objects.In this paper,the recognition and tracking stage and the CamShift algorithm are mainly written.This algorithm firstly calculates the HUE component histogram target under the HSV space,the probability distribution of the target pixel is obtained by reverse projection histogram,then call CV repository CamShift algorithm,automatic tracking and adjusting the center of the target window location and size.Experimental results show that the algorithm has good performance and can effectively track real-time target.

关 键 词：目标跟踪 OPENCV CAMSHIFT 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]