NTS-YOLO:a nocturnal traffic sign detection method based on improved YOLOv5

Author:

Zhang Yongchuan1,Guo Mengqi2,He Yong2,Xia Jun3,Geng Xuelai2,Zou Tao2,Ding Rui2

Affiliation:

1. Key Laboratory of Urban Land Resources Monitoring and Simulation, Ministry of Natural Resources

2. Smart City College, Chongqing Jiaotong University

3. Chongqing Institute of Surveying and Mapping Science and Technology

Abstract

Abstract

Accurate traffic sign recognition is one of the core technologies of intelligent driving systems, which face multiple challenges such as insufficient light and shadow interference at night. In this paper, we improve the YOLOv5 model for small, fuzzy, and partially occluded traffic sign targets at night and propose a high-precision nighttime traffic sign recognition method, "NTS-YOLO". The method firstly preprocesses the traffic sign dataset by adopting an unsupervised nighttime image enhancement method to improve the image quality under low-light conditions; secondly, it introduces the Convolutional Block Attention Module (CBAM) attentional mechanism, which focuses on the shape of the traffic sign by weighting the channel and spatial features inside the model and color to improve the perception under complex background and uneven illumination conditions; finally, the Optimal Transport Assignment (OTA) loss function is adopted to optimize the accuracy of predicting the bounding box and thus improve the performance of the model by comparing the difference between two probability distributions, i.e., minimizing the difference. In order to evaluate the effectiveness of the method, 154 samples of typical traffic signs containing small targets and fuzzy and partially occluded traffic signs with different lighting conditions in nighttime conditions were collected, and the data samples were subjected to CBMA, OTA, and a combination of the two methods, respectively, and comparative experiments were conducted with the traditional YOLOv5 algorithm. The experimental results show that "NTS-YOLO" achieves significant performance improvement in nighttime traffic sign recognition, with a mean average accuracy improvement of 0.95% for target detection of traffic signs and 0.17% for instance segmentation.

Publisher

Research Square Platform LLC

Reference38 articles.

1. Real-time color segmentation of road signs[C]//;Benallal M;IEEE Electrical and Computer Engineering,2003

2. Janssen R, Ritter W, Stein F, et al. Hybrid Approach For Traffic Sign Recognition[C]//IEEE Intelligent Vehicles' 93 Symposium, 2002:390–395.

3. Real-time traffic sign recognition from video by class-specific discriminative features[J];Ruta A;Pattern Recognition,2010

4. Ruta A, Porikli F, Watanabe S, et al. In-vehicle camera traffic sign detection and recognition[J]. Machine Vision & Applications, 2011, 22(2):359–375.

5. Real-Time Detection and Recognition of Road Traffic Signs[J];Greenhalgh J;IEEE Transactions on Intelligent Transportation Systems,2012

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3