基于OpenCV的视频运动目标检测与跟踪①.PDFVIP

计 算 机 系 统 应 用 2013 年 第 22 卷 第 5 期 ① 基于OpenCV 的视频运动目标检测与跟踪 1 1,2 1 袁宝红 , 张德祥 , 张玲君 1(安徽大学 电气工程与自动化学院, 合肥 230601) 2(安徽大学 计算智能与信号处理教育部重点实验室, 合肥 230601) 摘 要: 针对视频文件中运动目标检测与跟踪这一问题, 提出一种先检测后跟踪的方法. 首先利用平均背景法完 成对背景模型的更新, 从而检测出运动目标, 在此基础上利用投影法来投影出运动目标的大小, 最后再利用 MeanShift 算法对运动目标进行跟踪. 在跟踪过程中, 通过OpenCV 编写程序实现对运动目标的检测与跟踪. 实验 验证, 该方法在实现运动目标的精确检测与跟踪的基础上, 减少了运算量, 提高了跟踪的速度. 关键词: 目标检测; 跟踪; 平均背景法; MeanShift Detecting and Tracking of Moving Target in Video Sequence Based on OpenCV 1 1,2 1 YUAN Bao-Hong , ZHANG De-Xiang , ZHANG Ling-Jun 1(The Institute of Electrical Engineering and Automation, Anhui University, Hefei 230039, China) 2(Key Lab. of Intelligent Computing and Signal Processing of MOE, Anhui University, Hefei 230601, China) Abstract: In order to solve the moving target detecting and tracking in video sequence, the algorithm tracking follow detecting is proposed. First, it updated background model using the averaging background method, and detected the moving target. Then the size of the moving target was acquired by projection method. Finally, it tracked the moving target using the MeanShift tracking algorithm. During the object tracking, it programmed by the OpenCV to achieve the moving target detecting and tracking. The experiments show that the algorithm not only realize the detecting and tracking of the moving target exactly, but also reduce the amount of calculation and improve the tracking efficiency. Key words: target detecting; tracking; averaging background method; Meanshift 1 引言 小、形状和颜色等特性, 建立运动目标与前景目标间 视频中的目标检测与跟踪是指计算机代替人眼进 对应关系来实现跟踪.