摘要
准确检测并跟踪行人目标是进行人流量统计和行为分析的基础。在汽车客运站出入口监控系统中,多个行人一起行走的情况较多。针对这种情况,本文提出一种能分割多人的行人检测与跟踪方法。首先通过背景差法提取运动行人,并对多人融合的情况下进行行人分割;然后结合卡尔曼滤波原理跟踪行人,并输出其运动轨迹。实验结果表明,该方法能清楚地绘出行人的运动轨迹,具有很好的鲁棒性。
Accuracy pedestrian detection and tracking is the foundation of pedestrian counting and behavior analysis. In the intel- ligent video monitoring of bus passenger terminal, multiple pedestrians may walk together. A method of pedestrian detection and tracking, which can segment multiple pedestrians, is proposed. Firstly, pedestrians have been detected with the background sub- traction, and then multiple human objects have been segmented. Then the trajectory of pedestrians are shown after tracking pedes- trians with Kalman filter. The experimental results show that this method could draw the trajectory clearly, so as well robust re- sults.
出处
《计算机与现代化》
2013年第8期15-18,共4页
Computer and Modernization
基金
广州市科技计划重点项目(11A22050775)
关键词
汽车客运站
行人检测
行人分割
卡尔曼滤波
bus passenger terminal
pedestrian detection
pedestrian segmentation
Kanlman filter
