基于Hough变换检测前大灯的夜间车辆检测  

A Nighttime Vehicle Detection Method using Hough Transform to Detect Headlights

作　　者：许益成 谭文安[1] 陈丽婷[2] XU Yi-cheng;TAN Wen-an;Chen Li-ting(School of Computer Science and Technology,Nanjing University of Aeronautics and Astronautics,Nanjing 210016,China;School of Electrical and Information,Taizhou Vocational and Technical College,Taizhou 318000,China)

机构地区：[1]南京航空航天大学计算机科学与技术学院,南京210016 [2]台州职业技术学院电信学院,浙江台州318000

出　　处：《控制工程》2018年第8期1360-1363,共4页Control Engineering of China

基　　金：浙江省科技厅公益性项目(2014C33074);台州市科技计划项目(1202ky13)

摘　　要：针对夜间车辆检测与昼间车辆检测的不同,提出一种基于Hough圆检测的夜间车辆检测方法。首先利用灰度分割的方式提取车灯,对提取出的车灯进行连通域分析和边缘提取;然后进行Hough圆检测,得到车灯的半径和位置;最后利用车灯成对出现的特点,对车灯进行匹配确定车辆位置。实验结果表明,该算法对于夜间车辆检测检测率高,且虚警低。Aiming at the difference of nighttime vehicle detection and daytime vehicle detection, a nighttime vehicle detection method based on Hough circle detection is presented. First, extract the lamp using gray segmentation method, and carry on the connected domain analysis and edge extraction to extract the headlights; then use Hough circle detection to obtain the radius and position lights of the headlights; finally using the characteristics of lights appear in pairs, the lights are matched to determine the vehicle position. Experimental results show that this nighttime vehicle detection method has high detection rate and low false alarm.

关 键 词：HOUGH变换 圆检测 夜间车辆检测 车灯匹配 车灯检测 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]