基于红外成像的夜间车辆行驶轨迹识别方法  

Night vehicle trajectory recognition method based on infrared imaging

作　　者：刘小飞[1,2] 李明杰 LIU Xiaofei;LI Mingjie(School of Information&Intelligence Engineering,Sanya University,Sanya Hainan 572000,China;Academician Workstation of Guoliang Chen,Sanya Hainan 572000,China)

机构地区：[1]三亚学院信息与智能工程学院,海南三亚572000 [2]陈国良院士工作站,海南三亚572000

出　　处：《激光杂志》2022年第12期51-55,共5页Laser Journal

基　　金：海南省自然科学基金高层次人才项目(No.621RC1077);海南省自然科学基金资助项目(No.619MS076);海南省自然科学基金资助项目(No.620QN287);海南省自然科学基金资助项目(No.2019RC257)。

摘　　要：为提高夜间车辆行驶轨迹识别的准确性,设计一种基于红外成像的夜间车辆行驶轨迹识别方法。通过红外成像技术建立行驶轨迹采集模型,利用边缘轮廓检测方法识别红外特征,建立显著图特征分析模型,获取夜间交通视频图像序列,并检测车灯等信息,根据检测结果提取车辆行驶轨迹信息。实验结果表明,采用该方法进行夜间车辆行驶轨迹跟踪识别,在三侧车道的识别精度平均可达到95%以上,中间车道的最高识别精度可达到97%,识别时间平均为1 s。A night vehicle trajectory identification method based on infrared imaging is designed in order to improve the accuracy of night vehicle trajectory identification.Through infrared imaging technology,the driving track collection model is established.The infrared features are identified by an edge contour detection method.The feature analysis model of the saliency map is established,and the night traffic video image sequence is obtained.iInformation such as lights are detected,and the vehicle driving track information is extracted according to the detection results.The experimental results show that,when this method is used to track and identify the vehicles at night,the average recognition accuracy of the three lanes can reach more than 95%.The highest recognition accuracy of the middle lane can reach 97%,and the average recognition time is 1 s.

关 键 词：夜间车辆 行驶轨迹 红外成像 识别 边缘轮廓检测 

分 类 号：TN247[电子电信—物理电子学]