改进YOLOv5的交通标志检测算法

doi:10.3778/j.issn.1002-8331.2302-0319

摘要/Abstract

摘要： 如今对交通标志的检测在自动驾驶、智慧交通等领域是必不可少的关键环节，其关系到人们的驾驶安全问题。针对目前对交通标志的识别存在漏检、误检以及识别精度低、模型参数过多的问题，提出了一种基于改进YOLOv5的交通标志检测算法。增加一个小目标检测头，提高对小目标的识别精度。设计了一种由CBAM、SPConv、C3相结合的CSC3模块，引入YOLOv5主干网络中，同时减少其数量，目的是提升特征提取能力，降低参数量。将用于检测大目标的检测头删除，再把SPP替换成SPPCSPC，提高模型对交通标志的检测能力。增加跨层连接，并且通过重构Concat连接，目的是提高算法的识别精度。引入EIOU来替换CIOU损失函数，从而解决漏检、误检问题。使用DWConv替换主干网络的Conv，目的是减少模型参数，提高检测精度。实验结果表明，改进后的算法的平均准确率均值mAP@0.5：0.95为62.6%，比原YOLOv5s提高了8.3个百分点，参数量下降了10.1%，并且检测速度达到了74?FPS，能够满足实际检测需求。

关键词: YOLOv5, 交通标志检测, 特征融合, 深度可分离卷积, 目标检测

Abstract: Nowadays, the detection of traffic signs is an essential key link in automatic driving, intelligent transportation and other fields, which is related to people’s driving safety. A traffic sign detection algorithm based on improved YOLOv5 is proposed to solve the problems of missing detection, false detection, low recognition accuracy and excessive model parameters in the current traffic sign recognition. Firstly, a small target detection head is added to improve the recognition accuracy of small targets. Secondly, a CSC3 module combining CBAM, SPConv and C3 is designed, which is introduced into the YOLOv5 backbone network and reduces its number at the same time, in order to improve the feature extraction ability and reduce the number of parameters. The detection head used to detect large targets is deleted, and the SPP is replaced with SPPCSPC to improve the model’s ability to detect traffic signs. Cross-layer connections are added, and Concat connections are reconstructed to improve the recognition accuracy of the algorithm. EIOU is introduced to replace the CIOU loss function to solve the problem of missed detection and false detection. Finally, DWConv is used to replace the Conv of the backbone network to reduce the model parameters and improve the detection accuracy. The experimental results show that the average accuracy of the improved algorithm mAP @ 0.5：0.95 is 62.6%, which is 8.3 percentage points higher than the original YOLOv5s, the number of parameters has decreased by 10.1%, and the detection speed has reached 74 FPS, which can meet the actual detection requirements.

Key words: YOLOv5, traffic sign detection, feature fusion, depth-separable convolution, object detection

杨祥, 王华彬, 董明刚. 改进YOLOv5的交通标志检测算法[J]. 计算机工程与应用, 2023, 59(13): 194-204.

YANG Xiang, WANG Huabin, DONG Minggang. Improved YOLOv5’s Traffic Sign Detection Algorithm[J]. Computer Engineering and Applications, 2023, 59(13): 194-204.

参考文献

[1] REDMON J，DIVVALA S，GIRSHICK R，et al.You only look once：unified，real-time object detection[C]//Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition，2016：779-788.
[2] REDMON J，FARHADI A.YOLO9000：better，faster，stronger[J].arXiv：1612.08242，2016.
[3] REDMON J，FARHADI A.YOLOv3：an incremental improvement[J].arXiv：1804.02767，2018.
[4] BOCHKOVSKIY A，WANG C Y，LIAO H Y M.YOLOv4：optimal speed and accuracy of object detection[J].arXiv：2004.10934，2020.
[5] 李志刚，张娜.一种轻量型YOLOv5交通标志识别方法[J].电讯技术，2022，62（9）：1201-1206.
LI Z G，ZHANG N.A light-weight YOLOv5 traffic sign recongnition method[J].Telecommunication Engineering，2022，62（9）：1201-1206.
[6] MA N，ZHANG X，ZHENG H T，et al.Shufflenet v2：practical guidelines for efficient cnn architecture design[C]//Proceedings of the European Conference on Computer Vision（ECCV），2018：116-131
[7] ZHANG G，LI W J，ZHANG Y X.Traffic sign recognition based on the improved YOLOv5 algorithm[C]//Proceedings of the 2021 National Conference on Simulation Technology，2021：182-185.
[8] 李宇琼，周永军，蒋淑霞，等.基于注意力机制的交通标志识别[J].电子测量技术，2022，45（8）：116-120.
LI Y Q，ZHOU Y J，JIANG S X，et al.Traffic sign recognition based on attention mechanism[J].Electronic Measurement Technology，2022，45（8）：116-120.
[9] 王靖逸，刘树惠.基于改进YOLOv4的交通标志识别方法[J].电子设计工程，2022，30（18）：184-188.
WANG J Y，LIU S H.Traffic sign recognition method based on improved YOLOv4[J].Electronic Design Engineering，2022，30（18）：184-188.
[10] 党宏社，党晨，张选德.基于改进YOLOv5s的交通标志识别算法[J].实验技术与管理，2022，39（9）：97-102.
DANG H S，DANG C，ZHANG X D.Traffic sign recognition algorithm based on improved YOLOv5s[J].Experimental Technology and Management，2022，39（9）：97-102.
[11] 李松，亚森江·木沙.改进YOLOv7的X射线图像违禁品实时检测[J].计算机工程与应用，2023，59（12）：193-200.
LI S，MUSA Y S J.Improved YOLOv7 X-ray image real-time detection of prohibited items[J].Computer Engineering and Applications，2023，59（12）：193-200.
[12] 尹宋麟，谭飞，周晴，等.基于改进YOLOv4模型的交通标志检测[J].无线电工程，2022，52（11）：2087-2093.
YI S L，TAN F，ZHOU Q，et al.Traffic sign detection based on improved YOLOv4 model[J].Radio Engineering，2022，52（11）：2087-2093.
[13] 陈德海，孙仕儒，王昱朝，等.一种改进YOLOv3的交通标志识别算法[J].河南科技大学学报（自然科学版），2022，43（6）：31-36.
CHEN D H，SUN S R，WANG Y C，et al.An improved YOLOv3 traffic sign recognition algorithm[J].Journal of Henan University of Science and Technology（Natural Science），2022，43（6）：31-36.
[14] 秦强强，廖俊国，周弋荀.基于多分支混合注意力的小目标检测算法[J/OL].计算机应用：1-9（2023-03-16）[2023-03-24].http：//kns.cnki.net/kcms/detail/51.1307.TP.20230316.
1610.012.html.
QING Q Q，LIAO J G，ZHOU D X.Small object detection algorithm based on multi-branching hybrid attention[J/OL].Journal of Computer Applications：1-9（2023-03-16）[2023-03-24].http：//kns.cnki.net/kcms/detail/51.1307.
TP.20230316.1610.012.html.
[15] WANG J W，CHEN Y，GAO M Y，et al.Improved YOLOv5 network for real-time multi-scale traffic sign detection[J].arXiv：2112.08782，2021.
[16] 朱开，陈慈发.基于YOLOv5的雾霾天气下交通标志识别[J/OL].电子测量技术：1-8（2023-02-21）[2023-03-24].http：//kns.cnki.net/kcms/detail/11.2175.TN.20230221.1729.
010.html.
ZHU K，CHEN C F.Traffic sign recognition in hazy weather based on YOLOv5[J/OL].Electronic Measurement Techniques：1-8（2023-02-21）[2023-03-24].http：//kns.cnki.net/kcms/detail/11.2175.TN.20230221.1729.010.html.
[17] 郎斌柯，吕斌，吴建清，等.基于CA-BIFPN的交通标志检测模型[J/OL].深圳大学学报（理工版）：1-9（2023-02-23）[2023-03-24].http：//kns.cnki.net/kcms/detail/44.1401.N.20230223.1413.004.html.
LANG B K，LV B，WU J Q，et al.Traffic sign detection model based on CA-BIFPN[J/OL].Journal of Shenzhen University（Science and Technology Edition）：1-9（2023-02-23）[2023-03-24].http：//kns.cnki.net/kcms/detail/44.1401.N.20230223.1413.004.html.
[18] 胡均平，王鸿树，戴小标，等.改进YOLOv5的小目标交通标志实时检测算法[J].计算机工程与应用，2023，59（2）：185-193.
HU J P，WANG H S，DAI X B，et al.Real-time detection algorithm for small-target traffic signs based on improved YOLOv5[J].Computer Engineering and Applications，2023，59（2）：185-193.
[19] 徐正军，张强，许亮.一种基于改进YOLOv5s-Ghost网络的交通标志识别方法[J].光电子·激光，2023，34（1）：52-61.
XU Z J，ZHANG Q，XU L.A traffic sign identificatio method based on an improved YOLOv5s-Ghost network[J].Journal of Photoelectronics Laser，2023，34（1）：52-61.
[20] WANG C Y，BOCHKOVSKIY A，et al.YOLOv7：trainable bag-of-freebies sets new state-of-the-art for real-time object detectors[J].arXiv：2207.02696，2022.
[21] HE K，ZHANG X，REN S，et al.Deep residual learning for image recognition[C]//Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition，2016：770-778.
[22] HAH SAR，WU W J，LU Q M，et al.AmoebaNet：an SDN-enabled network service for big data science[J].Journal of Network and Computer Applications，2018，119：70-82.
[23] HOWARD A G，ZHU M，CHEN B，et al.Mobilenets：efficient convolutional neural networks for mobile vision applications[J].arXiv：1704.04861，2017.
[24] ZHANG X Y，ZHOU X Y，LIN M X，et al.ShuffleNet：an extremely efficient convolutional neural network for mobile devices[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition，2018：6848-6856.
[25] IANDOLA F N，HAN S，MOSKEWICZ M W，et al.SqueezeNet：AlexNet-level accuracy with 50x fewer parameters and <0.5?MB modedl size[J].arXiv：1602. 07360，2016.
[26] CHOLLEF F.Xception：deep learning with depthwise separable convolutions[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition，2017：1800-1807.
[27] SIMONYAN K Z A.Very deep convolutional networks for large-scale image recognition[J].arXiv：1409.1556，2014.
[28] LIN T Y，DOLLáR P，GIRSHICK R，et al.Feature pyramid networks for object detection[C]//Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition，2017：2117-2125
[29] LIU S，QI L，QIN H，et al.Path aggregation network for instance segmentation[C]//Proceedings of the 2018 IEEE Conference on Computer Vision and Pattern Recognition，2018：8759-8768.
[30] MAHENDRAN A，VEDALDI A.Understanding deep image representations by inverting them[C]//Proceedings of 2015 IEEE Conference on Computer Vision and Pattern Recognition，2015：5188-5196.
[31] HENDRYCKS D，GIMPEL K.Bridging nonlinearities and stochastic regularizers with gaussian error linear units[J].arXiv：1606.08415，2016.
[32] ZHANG J，HUANG M，JIN X，et al.A real-time Chinese traffic sign detection algorithm based on modified YOLOv2[J].Algorithms，2017，10（4）：127.