首页 | 本学科首页   官方微博 | 高级检索  
     检索      

速度异常的横穿马路行人检测算法
引用本文:许烨豪,毛琳,杨大伟.速度异常的横穿马路行人检测算法[J].大连民族学院学报,2018,20(3):218-221.
作者姓名:许烨豪  毛琳  杨大伟
作者单位:大连民族大学 机电工程学院,辽宁 大连 116605
基金项目:国家自然科学基金资助项目(61673084);辽宁省自然科学基金资助项目(20170540192)。
摘    要:针对人车混行区域行人异常运动极大地干扰驾驶者判断问题,提出一种单目摄像机条件下,基于TLD和粒子滤波器行人检测与跟踪技术的速度异常横穿马路行人检测方法,用于检测车前道路上的运动速度过快或过慢的行人目标,作为危险行人提醒驾驶者注意,从而实现行人目标的早期安全预警。经仿真分析,能够有效地估计实际道路上行人的运动速度,判断运动速度异常的行人目标,为驾驶者和车载辅助驾驶系统提供必要的参考信息。

关 键 词:目标检测  行人安全  速度异常  辅助驾驶  

Detection Algorithm of Jaywalkers with Velocity Abnormity
XU Ye-hao,MAO Lin,YANG Da-wei.Detection Algorithm of Jaywalkers with Velocity Abnormity[J].Journal of Dalian Nationalities University,2018,20(3):218-221.
Authors:XU Ye-hao  MAO Lin  YANG Da-wei
Institution:School of Electromechanical Engineering, Dalian Minzu University, Dalian Liaoning 116605, China
Abstract:Aiming at the problem that pedestrians′ abnormal movements in the mixed area of pedestrians and vehicles will greatly interfere the drivers′ judgment, this study applies monocular camera and proposes a detection algorithm of jaywalkers with velocity abnormity based on the detection and tracking method used TLD and particle filter. This method can detect the person who walks very fast or slow in front of vehicles, and warn drivers to achieve early protection for pedestrians. With the simulation, this algorithm effectively estimates pedestrian′s velocity on the road and detects the anomalous pedestrian, which provides necessary references to drivers and driver assistance systems.
Keywords:object detection  pedestrian safety  velocity abnormity  driver assistance  
点击此处可从《大连民族学院学报》浏览原始摘要信息
点击此处可从《大连民族学院学报》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号