基于车牌识别的机混车道视频测速算法  

Video speed measurement algorithm for vehicles in mixed roads based license plate recognition

作　　者：隋宗宾 高杨 梁宇[3] 王玉全[1] 

机构地区：[1]北方工业大学城市道路交通智能控制技术北京市重点实验室,北京100144 [2]北方工大学现场总线及自动化北京市重点实验室,北京100144 [3]中国石油天然气股份有限公司管道压缩机组维修中心,河北廊坊065000

出　　处：《现代电子技术》2015年第18期125-127,132,共4页Modern Electronics Technique

基　　金：学科建设-国家特殊需求-城市道路交通智能控制人才培养项目(市级)(PXM2015_014212_000023);研究生培养-产学研联合培养研究基地(级()PXM2015_014212_000018)

摘　　要：通过对视频测速应用场景的分析,提出一种在机混车道中应用视频测速的方法。该方法首先采用二维图像标定的方法获得视频坐标与实际地面坐标之间的映射关系,针对机混车道的复杂场景,采用Lab颜色空间和形态学处理相结合的方法对车牌进行识别,准确地跟踪到目标车辆,有效地避免了其他运动物体的干扰,同时提出了采用模板更新的方式防止目标车辆跟丢。最后通过适当选取车辆经过检测区时的图像帧,获得车辆的位移及对应的时间间隔,从而获得高精度的速度测量结果。Through analysis of video speed measurement scenarios,a way to detect the running speed of vehicles in mixed road by video speed measurement algorithm is proposed in this paper,in which the two-dimensional image labeling method is adopted to achieve the mapping relation between video coordinates and corresponding actual ground coordinates. Aiming at the complex scenarios of mixing road,the method of combining Lab color space with morphology is used to recognize the license plate,track the object vehicle precisely,avoid the disturbance of any other moving objects. Besides,the mode to refrain from missing object vehicles by a method of template updates is proposed. The displacement of vehicles and corresponding time inter-vals are got by picking up the image frame that the vehicles crossed the detection zone,with which the result of high-accuracy velocity measurement can be obtained.

关 键 词：视频测速 图像帧 车牌识别 目标跟踪 

分 类 号：TN941-34[电子电信—信号与信息处理] TP391.4[电子电信—信息与通信工程]