摘要
针对自动驾驶车辆识别红绿灯的需要,文章采用YOLO深度学习框架和Opencv视觉库,解决了交通灯识别的问题。试验结果表明,基于YOLO和Opencv的交通灯识别算法,能准确地识别出红灯、绿灯或黄灯,解决了自动驾驶车辆的交通灯识别难题。
In order to meet the need of traffic light recognition for autonomous vehicles,YOLO( a deep learning framework) and Opencv are adopted to solve the problem of traffic light recognition.The experimental results show that the traffic light recognition algorithm based on YOLO and Opencv can accurately identify the red,green or yellow lights,and solve the problem of autonomous vehicle traffic light recognition.
引文
[1]宋凯凯,汪涛.基于视觉的前方车辆与交通灯识别[J].汽车工业研究,2018(12):54-59.
[2]金充充,何立平,朱仲杰,等.雾霾天气下交通信号灯定位与识别算法[J].浙江万里学院学报,2018,31(6):61-68.
[3]朱永珍,孟庆虎,普杰信.基于HSV色彩空间与形状特征的交通灯自动识別[J].电视技术,2015,39(5):150-154.
[4]李广亮,杨成忠,张泽泓.智能车交通灯识别[J].杭州电子科技大学学报,2014,34(3):80-82.
[5]叶茂胜,李必军,莫柠锴,等.基于色域差分与伽马校正的交通灯识别[J].软件导刊,2016,15(9):157-160.
[6]贾瑞明,刘立强.基于卷积神经网络的交通灯识别[J].北方工业大学学报,2018,30(5):22-30.
[7]孙铁强,贺颖.基于Faster R-CNN的辅助驾驶相关检测[J].华北理工大学学报(自然科学版),2018,40(4):112-119.