计算机科学 ›› 2019, Vol. 46 ›› Issue (12): 272-278.doi: 10.11896/jsjkx.190400026

• 图形图像与模式识别 • 上一篇    下一篇

基于深度学习的交通信号灯快速检测与识别

钱弘毅1, 王丽华1, 牟宏磊2   

  1. (北京航空航天大学软件学院 北京100191)1;
    (西北工业大学自动化学院 西安710072)2
  • 收稿日期:2019-04-08 出版日期:2019-12-15 发布日期:2019-12-17
  • 通讯作者: 王丽华(1961-),女,教授,主要研究方向为集成电路与物联网、软件工程项目管理与开发、基于视觉的机器人,E-mail:wanglihua@buaa.edu.cn。
  • 作者简介:钱弘毅(1993-),男,硕士生,主要研究方向为深度学习、无人驾驶,E-mail:13861408913@163.com;牟宏磊(1981-),男,硕士生,主要研究方向为无人系统。

Fast Detection and Identification of Traffic Lights Based on Deep Learning

QIAN Hong-yi1, WANG Li-hua1, MOU Hong-lei2   

  1. (College of Software,Beihang University,Beijing 100191,China)1;
    (School of Automation,Northwestern Polytechnical University,Xi’an 710072,China)2
  • Received:2019-04-08 Online:2019-12-15 Published:2019-12-17

摘要: 交通信号灯检测与识别技术能够辅助司机做出正确的驾驶决策,减少交通事故的发生,为无人驾驶的实现提供安全保障。针对交通信号灯检测场景复杂多变、目标通常占检测数据集图片的比例极小等技术难点,提出了一种基于深度学习的交通信号灯快速检测与识别算法。整体框架包括如下3部分:基于启发式的图像预分割,用于缩小搜索范围,提升信号灯面板在输入图像中的相对大小和检测精度;基于深度学习的检测与识别,利用卷积神经网络准确地检测与识别信号灯;利用NMS(Non-Maximum Suppression)算法去除上一阶段中重复的检测框。提出的Split-CS-Yolo模型在LISA数据集上取得了96.08%的mAP和2.87%的漏检率,相比Yolo系列的其他方法,其不仅有更高的准确率和更低的漏检率,还将模型大小缩小到原始Yolov2的8.6%,使得检测速度提升了63%。

关键词: NMS, 交通信号灯检测与识别, 快速检测, 深度学习, 图像预分割

Abstract: Traffic light detection and recognition technology can help drivers make correct driving decisions,reduce traffic accidents,and provide security for unmanned driving.Aiming at the technical difficulties such as the complex and variable traffic light detection scene,and targets typically account for a very small percentage of the dataset images,a fast detection and recognition algorithm for traffic light based on deep learning was proposed.The overall framework consists of three parts:heuristic-based image pre-segmentation,which is used to narrow the search range and improve the relative size and detection accuracy of the traffic light panel in the input images;detection and recognition based on deep learning,using convolutional neural networks to detect and identify traffic lights accurately;NMS (Non-Maximum Suppression) algorithm,which is used to remove the repeated detections of the previous stage.The proposed Split-CS-Yolo model achieves 96.08% mAP and 2.87% miss detection rate on the LISA dataset.Compared with other methods of the Yolo series,it not only has higher accuracy and lower missed detection rate,but also reduces the model size to 8.6% of the original Yolov2,thus increasing the detection speed by 63%.

Key words: Deep learning, Fast detection, Image pre-segmentation, NMS, Traffic light detectionand recognition

中图分类号: 

  • TP391
[1]XU M W,ZHANG C Y.Traffic Light Detection and Recognition Based on Salience Map[J].Computer and Digital Enginee-ring,2017,45(7):1397-1401.(in Chinese)
许明文,张重阳.基于显著性特征的交通信号灯检测和识别[J].计算机与数字工程,2017,45(7):1397-1401.
[2]KOCH C,ULLMAN S.Shifts in selective visual attention:towards the underlying neural circuitry[J].Hum Neurobiol,1987,4(4):219-227.
[3]SHI X M,ZHAO N,XIA Y J.Detection and classification of traffic lights for automated setup of road surveillance systems[J].Multimedia Tools & Applications,2016,75(20):12547-12562.
[4]JI Y,YANG M,LU Z,et al.Integrating visual selective attention model with HOG features for traffic light detection and recognition[C]//Intelligent Vehicles Symposium.Korea:IEEE,2015:280-285.
[5]ZHANG Q R,GU G C,XIAO H M.Computational Model of Visual Selective Attention[J].Robot,2009,31(6):574-580.
[6]YANG M,LU Z C,GUO L D,et al.Vision-Based Environmental Perception and Navigation of Micro-Intelligent Vehicles[C]//Advances in Intelligent Systems and Computing.2014:653-665.
[7]LIU Q,QIN S Y,LI Z C.The Modeling of Visual Selective Attention and Its Applications[J].Science & Technology Review,2010,28(1):107-115.(in Chinese)
刘琼,秦世引,李志成.视觉选择性注意的模型化计算及其应用前景[J].科技导报,2010,28(1):107-115.
[8]JANG C H,KIM C,KIM D,et al.Multiple exposure images based traffic light recognition[C]//IEEE Intelligent Vehicles Symposium Proceedings.USA:IEEE,2014:1313-1318.
[9]OMACHI M,OMACHI S.Detection of traffic light using structural information[C]//IEEE 10th International Conference on Signal Processing.Beijing:IEEE,2010:809-812.
[10]KIM Y K,KIM K W,YANG X L.Real Time Traffic Light Re- cognition System for Color Vision Deficiencies[C]//Internatio-nal Conference on Mechatronics and Automation.Harbin:IEEE,2007:76-81.
[11]ZHANG Y,XUE J,ZHANG G,et al.A multi-feature fusion based traffic light recognition algorithm for intelligent vehicles[C]//Proceedings of the 33rd Chinese Control Conference.2014:4924-4929.
[12]KASTNER R,MICHALKE T,BURBACH T,et al.Attention-based traffic sign recognition with an array of weak classifiers[C]//Intelligent Vehicles Symposium.San Deigo,CA:IEEE,2010:333-339.
[13]LI Z K,XU G Z,GUO M M.Traffic signal recognition method based on gradient content feature[J].Transducer and Microsystem Technologies,2018,37(10):38-40.(in Chinese)
李子康,徐桂芝,郭苗苗.基于渐变上下文特征的交通灯识别方法[J].传感器与微系统,2018,37(10):38-40.
[14]KIM J,CHO H,HWANGBO M,et al.Deep Traffic Light Detection for Self-driving Cars from a Large-scale Dataset[C]//IEEE 2018 21st International Conference on Intelligent Transportation Systems (ITSC).USA:IEEE,2018:280-285.
[15]LIU W,ANGUELOV D,ERHAN D,et al.SSD:Single shot multibox detector[C]//Proceedings of European Conference on Computer Vision.2016:21-37.
[16]MÜLLER J,DIETMAYER K.Detecting Traffic Lights by Single Shot Detection[C]//IEEE International Conference on Intelligent Transportation Systems (ITSC).IEEE,2018.
[17]SZEGEDY C,VANHOUCKE V,IOFFE S,et al.Rethinking the Inception Architecture for Computer Vision[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).IEEE,2016:2818-2826.
[18]SIMONYAN K,ZISSERMAN A.Very deep convolutional networks for large-scale image recognition[C]//Proceedings of ICLR.2015.
[19]HAN C,GAOG Y,ZHANG Y.Real-time small traffic sign detection with revised faster-RCNN[J].Multimedia Tools and Applications,2018,78(10):13263-13278.
[20]REN S,HE K,GIRSHICK R,et al.Faster R-CNN:Towards Real-Time Object Detection with Region Proposal Networks[J].IEEE Transactions on Pattern Analysis & Machine Intelligence,2017,39(6):1137-1149.
[21]LU Y F,LU J M,ZHANG S H,et al.Traffic signal detection and classification in street views using an attention model[J].Computational Visual Media,2018,4(3):253-266.
[22]NIU J W,LIU Y,GUIZANI M,et al.Deep CNN-based Real-time Traffic Light Detector for Self-driving Vehicles[C]//IEEE Transactions on Mobile Computing.2019.
[23]PHILIPSEN M P,JENSEN M B,MOGELMOSE A,et al.Traffic light detection:a learning algorithm and evaluations oncha-llenging dataset[C]//2015 IEEE 18th International Conference on Intelligent Transportation Systems (ITSC).IEEE,2015:2341-2345.
[24]REDMON J,DIVVALA S,GIRSHICK R,et al.You only look once:Unified,real-time object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2016:779-788.
[25]REDMONJ,FARHADI A.YOLO9000:Better,Faster,Stronger[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).IEEE,2017:6517-6525.
[26]MA N,ZHANG X,ZHENG H T,et al.ShuffleNet V2:Practical Guidelines for Efficient CNN Architecture Design[C]//Lecture Notes in Computer Science(ECCV 2018).Cham:Springer,2018.
[27]IOFFE S,SZEGEDY C.Batch Normalization:Accelerating Deep Network Training by Reducing Internal Covariate Shift[C]//Proceedings of the 32nd International Conference on International Conference on Machine Learning(ICML’15).2015:448-456.
[28]REDMON J,FARHADI A.YOLOv3:An Incremental Improvement[J].arXiv:1804.02767.
[29]YOSINSKI J,CLUNE J,BENGIO Y,et al.How transferable are features in deep neural networks?[C]//Proceedings of the 27th International Conference on Nueral Information Processing Systems.2014:3320-3328.
[30]LU Y,LU J,ZHANG S,et al.Traffic signal detection and classification in street views using an attention model[J].Computational Visual Media,2018,4(3):59-72.
[1] 饶志双, 贾真, 张凡, 李天瑞.
基于Key-Value关联记忆网络的知识图谱问答方法
Key-Value Relational Memory Networks for Question Answering over Knowledge Graph
计算机科学, 2022, 49(9): 202-207. https://doi.org/10.11896/jsjkx.220300277
[2] 汤凌韬, 王迪, 张鲁飞, 刘盛云.
基于安全多方计算和差分隐私的联邦学习方案
Federated Learning Scheme Based on Secure Multi-party Computation and Differential Privacy
计算机科学, 2022, 49(9): 297-305. https://doi.org/10.11896/jsjkx.210800108
[3] 徐涌鑫, 赵俊峰, 王亚沙, 谢冰, 杨恺.
时序知识图谱表示学习
Temporal Knowledge Graph Representation Learning
计算机科学, 2022, 49(9): 162-171. https://doi.org/10.11896/jsjkx.220500204
[4] 王剑, 彭雨琦, 赵宇斐, 杨健.
基于深度学习的社交网络舆情信息抽取方法综述
Survey of Social Network Public Opinion Information Extraction Based on Deep Learning
计算机科学, 2022, 49(8): 279-293. https://doi.org/10.11896/jsjkx.220300099
[5] 郝志荣, 陈龙, 黄嘉成.
面向文本分类的类别区分式通用对抗攻击方法
Class Discriminative Universal Adversarial Attack for Text Classification
计算机科学, 2022, 49(8): 323-329. https://doi.org/10.11896/jsjkx.220200077
[6] 姜梦函, 李邵梅, 郑洪浩, 张建朋.
基于改进位置编码的谣言检测模型
Rumor Detection Model Based on Improved Position Embedding
计算机科学, 2022, 49(8): 330-335. https://doi.org/10.11896/jsjkx.210600046
[7] 孙奇, 吉根林, 张杰.
基于非局部注意力生成对抗网络的视频异常事件检测方法
Non-local Attention Based Generative Adversarial Network for Video Abnormal Event Detection
计算机科学, 2022, 49(8): 172-177. https://doi.org/10.11896/jsjkx.210600061
[8] 侯钰涛, 阿布都克力木·阿布力孜, 哈里旦木·阿布都克里木.
中文预训练模型研究进展
Advances in Chinese Pre-training Models
计算机科学, 2022, 49(7): 148-163. https://doi.org/10.11896/jsjkx.211200018
[9] 周慧, 施皓晨, 屠要峰, 黄圣君.
基于主动采样的深度鲁棒神经网络学习
Robust Deep Neural Network Learning Based on Active Sampling
计算机科学, 2022, 49(7): 164-169. https://doi.org/10.11896/jsjkx.210600044
[10] 苏丹宁, 曹桂涛, 王燕楠, 王宏, 任赫.
小样本雷达辐射源识别的深度学习方法综述
Survey of Deep Learning for Radar Emitter Identification Based on Small Sample
计算机科学, 2022, 49(7): 226-235. https://doi.org/10.11896/jsjkx.210600138
[11] 胡艳羽, 赵龙, 董祥军.
一种用于癌症分类的两阶段深度特征选择提取算法
Two-stage Deep Feature Selection Extraction Algorithm for Cancer Classification
计算机科学, 2022, 49(7): 73-78. https://doi.org/10.11896/jsjkx.210500092
[12] 程成, 降爱莲.
基于多路径特征提取的实时语义分割方法
Real-time Semantic Segmentation Method Based on Multi-path Feature Extraction
计算机科学, 2022, 49(7): 120-126. https://doi.org/10.11896/jsjkx.210500157
[13] 王君锋, 刘凡, 杨赛, 吕坦悦, 陈峙宇, 许峰.
基于多源迁移学习的大坝裂缝检测
Dam Crack Detection Based on Multi-source Transfer Learning
计算机科学, 2022, 49(6A): 319-324. https://doi.org/10.11896/jsjkx.210500124
[14] 楚玉春, 龚航, 王学芳, 刘培顺.
基于YOLOv4的目标检测知识蒸馏算法研究
Study on Knowledge Distillation of Target Detection Algorithm Based on YOLOv4
计算机科学, 2022, 49(6A): 337-344. https://doi.org/10.11896/jsjkx.210600204
[15] 周志豪, 陈磊, 伍翔, 丘东亮, 梁广升, 曾凡巧.
基于SMOTE-SDSAE-SVM的车载CAN总线入侵检测算法
SMOTE-SDSAE-SVM Based Vehicle CAN Bus Intrusion Detection Algorithm
计算机科学, 2022, 49(6A): 562-570. https://doi.org/10.11896/jsjkx.210700106
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!