Traffic light detection and recognition based on deep learning for autonomous-rail rapid tram

The autonomous rail rapid transit (ART) system is an intelligent express transport solution developed independently by CRRC Zhuzhou institute co., ltd. The detection and recognition of traffic lights are critical for improving the safety of the ART automatic operation system. However, existing means...

Full description

Saved in:
Bibliographic Details
Main Authors: XIONG Qunfang, LIN Jun, YUAN Xiwen, XU Yanghan, YUE Wei, LI Yuanzhengyu
Format: Article
Language:zho
Published: Editorial Department of Electric Drive for Locomotives 2024-11-01
Series:机车电传动
Subjects:
Online Access:http://edl.csrzic.com/thesisDetails#10.13890/j.issn.1000-128X.2024.01.244
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849728888022761472
author XIONG Qunfang
LIN Jun
YUAN Xiwen
XU Yanghan
YUE Wei
LI Yuanzhengyu
author_facet XIONG Qunfang
LIN Jun
YUAN Xiwen
XU Yanghan
YUE Wei
LI Yuanzhengyu
author_sort XIONG Qunfang
collection DOAJ
description The autonomous rail rapid transit (ART) system is an intelligent express transport solution developed independently by CRRC Zhuzhou institute co., ltd. The detection and recognition of traffic lights are critical for improving the safety of the ART automatic operation system. However, existing means for traffic light detection and recognition often fall short of meeting the detection requirements in the automatic operation environments characterized by customized traffic lights, with only a few regular signals. This paper presents study efforts in this field through the application of a deep learning algorithm. Firstly, regions of interest (RoIs) for traffic lights were determined using high-precision map information to narrow the detection range and improve the detection speed. Secondly, an improved YOLOV5s network was employed to extract features from the RoIs, facilitating the recognition of ART traffic lights. Finally, the extracted traffic signal images were classified using a MobileNetV2 lightweight network to identify specific signal categories. In order to further enhance the model's generalization performance, an image diagnosis algorithm was introduced before signal recognition, which generates warnings for complex conditions, such as overexposure and backlighting. These abnormal image data were saved and utilized for further training and optimizing the model. Experimental results reveals that the proposed approach effectively detected and recognized ART traffic lights, achieving an average detection precision of 84.76% on designated roads during the daytime while exhibiting good real-time performance.
format Article
id doaj-art-b734af55acc24d3c869da082e70e55c7
institution DOAJ
issn 1000-128X
language zho
publishDate 2024-11-01
publisher Editorial Department of Electric Drive for Locomotives
record_format Article
series 机车电传动
spelling doaj-art-b734af55acc24d3c869da082e70e55c72025-08-20T03:09:25ZzhoEditorial Department of Electric Drive for Locomotives机车电传动1000-128X2024-11-0114014856193321Traffic light detection and recognition based on deep learning for autonomous-rail rapid tramXIONG QunfangLIN JunYUAN XiwenXU YanghanYUE WeiLI YuanzhengyuThe autonomous rail rapid transit (ART) system is an intelligent express transport solution developed independently by CRRC Zhuzhou institute co., ltd. The detection and recognition of traffic lights are critical for improving the safety of the ART automatic operation system. However, existing means for traffic light detection and recognition often fall short of meeting the detection requirements in the automatic operation environments characterized by customized traffic lights, with only a few regular signals. This paper presents study efforts in this field through the application of a deep learning algorithm. Firstly, regions of interest (RoIs) for traffic lights were determined using high-precision map information to narrow the detection range and improve the detection speed. Secondly, an improved YOLOV5s network was employed to extract features from the RoIs, facilitating the recognition of ART traffic lights. Finally, the extracted traffic signal images were classified using a MobileNetV2 lightweight network to identify specific signal categories. In order to further enhance the model's generalization performance, an image diagnosis algorithm was introduced before signal recognition, which generates warnings for complex conditions, such as overexposure and backlighting. These abnormal image data were saved and utilized for further training and optimizing the model. Experimental results reveals that the proposed approach effectively detected and recognized ART traffic lights, achieving an average detection precision of 84.76% on designated roads during the daytime while exhibiting good real-time performance.http://edl.csrzic.com/thesisDetails#10.13890/j.issn.1000-128X.2024.01.244deep learningART traffic lighthigh-precision mapYOLOV5sMobileNetV2image diagnosis
spellingShingle XIONG Qunfang
LIN Jun
YUAN Xiwen
XU Yanghan
YUE Wei
LI Yuanzhengyu
Traffic light detection and recognition based on deep learning for autonomous-rail rapid tram
机车电传动
deep learning
ART traffic light
high-precision map
YOLOV5s
MobileNetV2
image diagnosis
title Traffic light detection and recognition based on deep learning for autonomous-rail rapid tram
title_full Traffic light detection and recognition based on deep learning for autonomous-rail rapid tram
title_fullStr Traffic light detection and recognition based on deep learning for autonomous-rail rapid tram
title_full_unstemmed Traffic light detection and recognition based on deep learning for autonomous-rail rapid tram
title_short Traffic light detection and recognition based on deep learning for autonomous-rail rapid tram
title_sort traffic light detection and recognition based on deep learning for autonomous rail rapid tram
topic deep learning
ART traffic light
high-precision map
YOLOV5s
MobileNetV2
image diagnosis
url http://edl.csrzic.com/thesisDetails#10.13890/j.issn.1000-128X.2024.01.244
work_keys_str_mv AT xiongqunfang trafficlightdetectionandrecognitionbasedondeeplearningforautonomousrailrapidtram
AT linjun trafficlightdetectionandrecognitionbasedondeeplearningforautonomousrailrapidtram
AT yuanxiwen trafficlightdetectionandrecognitionbasedondeeplearningforautonomousrailrapidtram
AT xuyanghan trafficlightdetectionandrecognitionbasedondeeplearningforautonomousrailrapidtram
AT yuewei trafficlightdetectionandrecognitionbasedondeeplearningforautonomousrailrapidtram
AT liyuanzhengyu trafficlightdetectionandrecognitionbasedondeeplearningforautonomousrailrapidtram