Research on Recognition of Highway Toll Vehicles Based on Dual-View Feature Fusion Network

Authors

• XIAO Qiang, School of Information Science and Engineering of Yunnan University , Kunming , Yunnan 650011 , China ;Yunnan Highway Network Toll Management Co ., Ltd., Kunming , Yunnan 650011 , China ;Yunnan Key Laboratory of Digital Communications , Kunming , Yunnan 650011 , ChinaFollow
• CHEN Xinjia, School of Information Science and Engineering of Yunnan University , Kunming , Yunnan 650011 , ChinaFollow
• CHEN Jing, Yunnan Highway Network Toll Management Co ., Ltd., Kunming , Yunnan 650011 , China ;Yunnan Key Laboratory of Digital Communications , Kunming , Yunnan 650011 , China
• ZHAO Bin, Yunnan Highway Network Toll Management Co ., Ltd., Kunming , Yunnan 650011 , China ;Yunnan Key Laboratory of Digital Communications , Kunming , Yunnan 650011 , China
• YAN Kaifeng, Yunnan Highway Network Toll Management Co ., Ltd., Kunming , Yunnan 650011 , China
• DING Hongwei, School of Information Science and Engineering of Yunnan University , Kunming , Yunnan 650011 , China

Corresponding Author

陈心嘉，女，硕士研究生.E-mail：2240821337@qq.com

Abstract

To solve the practical problems in vehicle type classification on toll roads, a deep learning network based on dual-view feature fusion, named ETCLNet, was proposed in this paper. A dual-branch architecture was adopted by the network to extract the features of vehicle head and side body images, respectively, and combined with an innovatively designed multi-scale feature extraction module, namely IPFE, and an adaptive weighting-based bilinear feature fusion (AWBF) mechanism, efficient and fine-grained vehicle feature representation was achieved. Through the design of parallel multi-scale convolutions and residual connections, the problem of gradient vanishing was effectively alleviated, and the adaptability to complex scenarios was enhanced by the IPFE module. The high-order interaction of features between angles was further optimized by the AWBF mechanism, and efficient fusion of multi-angle features was achieved, and classification performance was significantly improved through adaptive weighting and element-wise multiplication. Experimental results indicate that ETCLNet outperforms existing mainstream models in multiple evaluation metrics (e.g., accuracy, precision, recall, F1 value, and SAUC), which solves the problem of the low vehicle type recognition rate of cameras in existing toll road audit systems. In addition, existing hardware devices are fully utilized by this design, which reduces deployment costs. A scientific and efficient solution for multi-view vehicle recognition and smart transportation is provided by this paper, and new ideas for deep learning network design are offered.

Publication Date

6-27-2026

DOI

10.14048/j.issn.1671-2579.2026.03.026

First Page

238

Last Page

247

Submission Date

June 2026

Recommended Citation

Qiang, XIAO; Xinjia, CHEN; Jing, CHEN; Bin, ZHAO; Kaifeng, YAN; and Hongwei, DING (2026) "Research on Recognition of Highway Toll Vehicles Based on Dual-View Feature Fusion Network," Journal of China & Foreign Highway: Vol. 46: Iss. 3, Article 26.
DOI: 10.14048/j.issn.1671-2579.2026.03.026
Available at: https://zwgl1980.csust.edu.cn/journal/vol46/iss3/26

Reference

[1]匡永强,胡先全.基于多数据融合的高速公路收费稽核系统设计 [J].中国交通信息化,2022 (增刊 1):254-256，278.KUANG Yongqiang,HU Xianquan.Design of expressway toll audit system based on multi-data fusion [J].China ITS Journal,2022 (sup 1):254-256,278.
[2]史元超.基于新型传感器的交通状态获取技术的研究[D].北京:北京交通大学,2009.SHI Yuanchao.Research on traffic state acquisition technology based on new sensor [D].Beijing:Beijing Jiaotong University,2009.
[3]彭堉斌.基于 FMCW毫米波雷达的车辆轨迹跟踪和车型识别系统 [D].哈尔滨:哈尔滨工业大学,2020.PENG Yubin.Vehicle trajectory tracking and vehicle identification system based on FMCW millimeter wave radar [D].Harbin:Harbin Institute of Technology,2020.
[4]王君逸.基于多特征融合的交通车辆声信号识别方法研究[D].南昌:华东交通大学,2022.WANG Junyi.Research on traffic vehicle acoustic signal recognition method based on multi-feature fusion [D].Nanchang:East China Jiaotong University,2022.
[5]柳长源,何先平,毕晓君.融合注意力机制的高效率网络车 型 识 别 [J].浙 江 大 学 学 报 (工 学 版 ),2022,56(4):775-782.LIU Changyuan,HE Xianping,BI Xiaojun.Efficient network vehicle recognition combined with attention mechanism [J].Journal of Zhejiang University (Engineering Science ),2022,56(4):775-782.
[6]曾夏明,何家雄,曹林根,等.基于激光雷达车检器的车型 识 别 研 究 [J].中 国 工 程 机 械 学 报,2021,19(4):324-330.ZENG Xiaming,HE Jiaxiong,CAO Lingen,et al.Research on vehicle recognition based on laser radar vehicle inspector [J].Chinese Journal of Construction Machinery,2021,19(4):324-330.
[7]郑健龙,陈梦洁,刘超超.公路基础设施智能建造发展与展望综述 [J].中外公路,2025,45(2):1-20.ZHENG Jianlong,CHEN Mengjie,LIU Chaochao.Review of intelligent construction development of highway infrastructures and its prospects [J].Journal of China & Foreign Highway,2025,45(2):1-20.
[8]邓航.基于 Gabor特征简化的车型识别 [D].西安:西安电子科技大学,2017.DENG Hang.Vehicle recognition based on Gabor feature simplification [D].Xi’an:Xidian University,2017.
[9]罗亮,吕俊杰,李涛,等.基于改进的 NC-HOG 特征的工程车车型自动识别算法 [J].计算机工程与设计,2021,42(11):3164 -3173.LUO Liang,LYU Junjie,LI Tao,et al.Automatic recognition of engineering vehicle type based on improved NC-HOG feature [J].Computer Engineering and Design,2021,42(11):3164 -3173.
[10]BISHARA A,MAZE E H,MAZE M.Considerations for the implementation of machine learning into acute care settings [J].British Medical Bulletin,2022,141(1):15-32.
[11]ZHAO P N,LAI L F.Minimax rate optimal adaptive nearest neighbor classification and regression [J].IEEE Transactions on Information Theory,2021,67(5):3155 -3182.
[12]宋嘉冀.基于视频的车辆信息提取与检索技术研究 [D].南京:东南大学,2017.SONG Jiaji.Research on vehicle information extraction and retrieval technology based on video [D].Nanjing:Southeast University,2017.
[13]程淑红,高许,周斌.基于多特征提取和 SVM参数优化的车型识别 [J].计量学报,2018,39(3):348-352.CHENG Shuhong,GAO Xu,ZHOU Bin.Vehicle recognition based on multi-feature extraction and SVM parameter optimization [J].Acta Metrologica Sinica,2018,39(3):348-352.
[14]LECUN Y,BENGIO Y,HINTON G.Deep learning [J].Nature,2015,521(7553 ):436-444.
[15]YAO D H,ZHU W X,CHEN Y J,et al.Chinese license plate character recognition based on convolution neural network [C]//Jinan:2017 Chinese Automation Congress (CAC ).2017:1547 -1552.
[16]何梓林.基于 Tensorflow 的车型识别系统研究 [J].农业装备与车辆工程,2022,60(8):135-138.HE Zilin.Research on vehicle recognition system based on Tensorflow [J].Agricultural Equipment & Vehicle Engineering,2022,60(8):135-138.
[17]牟亮,赵红,李燕,等.基于梯度压缩的 YOLO v 4算法车型识别 [J].工程科学学报,2022,44(5):940-950.MU Liang,ZHAO Hong,LI Yan,et al.Vehicle recognition based on gradient compression and YOLO v 4 algorithm [J].Chinese Journal of Engineering,2022,44(5):940-950.
[18]张念,张亮.基于深度学习的公路货车车型识别 [J].交通运输工程学报,2023,23(1):267-279.ZHANG Nian,ZHANG Liang.Type recognition of highway trucks based on deep learning [J].Journal of Traffic and Transportation Engineering,2023,23(1):267-279.
[19]封慧杰,赵红东,于快快,等.基于轻量级卷积神经网络的 车 辆 声 学 识 别 [J].传 感 器 与 微 系 统,2024,43(7):136-140.FENG Huijie,ZHAO Hongdong,YU Kuaikuai,et al.Vehicle acoustic recognition based on lightweight CNN[J].Transducer and Microsystem Technologies,2024,43(7):136-140.
[20]徐胜军,荆扬,李海涛,等.渐进式多粒度 ResNet车型识别网络 [J].光电工程,2023,50(7):36-51.XU Shengjun,JING Yang,LI Haitao,et al.Progressive multi-granularity ResNet vehicle recognition network [J].Opto-Electronic Engineering,2023,50(7):36-51.
[21]江屾.基于改进 YOLO v 5的车辆检测及跟踪方法研究[D].重庆:重庆交通大学,2023.JIANG Shen.Research on vehicle detection and tracking method based on improved YOLO v 5[D].Chongqing:Chongqing Jiaotong University,2023.
[22]王岱.基于改进 YOLOX 的恶劣天气车道线与车辆检测识别 [D].北京:中国地质大学 (北京 ),2023.WANG Dai.Detection and identification of vehicle lanes and vehicles in bad weather based on improved YOLOX[D].Beijing:China University of Geosciences,2023.
[23]王开明,罗哲皓,孙文琪,等.基于 FIPA模型的自适应协同优化的车型识别算法 [J].计算机应用,2024,44(增刊 2):267-273.WANG Kaiming,LUO Zhehao,SUN Wenqi,et al.Vehicle recognition algorithm based on FIPA model and adaptive collaborative optimization [J].Journal of Computer Applications,2024,44(sup 2):267-273.
[24]董一兵,曾辉,李建科,等.DPRT-YOLO:智能网联汽车复杂驾驶环境实时目标检测器 [J].计算机工程与应用,2025,61(14):148-162.DONG Yibing,ZENG Hui,LI Jianke,et al.DPRT-YOLO:Real-time object detector for intelligent and connected vehicles in complex driving environments [J].Computer Engineering and Applications,2025,61(14):148-162.
[25]曾永强,张永华,赵辉,等.高速路侧雷达和相机融合的目标检测与跟踪 [J].中外公路,2025,45(6):261-267.ZENG Yongqiang,ZHANG Yonghua,ZHAO Hui,et al.Object detection and tracking based on fusion of radar and camera on highway side [J].Journal of China & Foreign Highway,2025,45(6):261-267.
[26]VASWANI A,SHAZEER N,PARMAR N,et al.Attention is all you need [PP/OL ].arXiv.(2011 -06-03) [2025 -02-12].https://arxiv.org/abs/ 1706.03762.
[27]SZEGEDY C,LIU W,JIA Y Q,et al.Going deeper with convolutions [C]//Boston:2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR ),2015:1-9.
[28]SZEGEDY C,VANHOUCKE V,IOFFE S,et al.Rethinking the inception architecture for computer vision[C]//Las Vegas:2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR ),2016:2818 -2826.
[29]SANDLER M,HOWARD A,ZHU M L,et al.MobileNet v2:Inverted residuals and linear bottlenecks [C]//Salt Lake City：2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition，2018:4510 -4520.
[30]KOONCE B.Convolutional Neural networks with swift for tensorflow:Image recognition and dataset categorization [M].Berkeley,CA:Apress,2021.
[31]IANDOLA F N,HAN S,MOSKEWICZ M W,et al.SqueezeNet:AlexNet-level accuracy with 50x fewer parameters and < 0.5 MB model size [PP/OL ].arXiv.(2016 -02-07)[2025 -02-12].https://arxiv.org/abs/ 1602.07360.
[32]DAI Y Q,ZHENG T,XUE C B,et al.MViT-PCD:A lightweight ViT-based network for Martian surface topographic change detection [J].IEEE Geoscience and Remote Sensing Letters,2023,20:3000405.
[33]WANG C Y,YEH I H,LIAO H M.YOLO v 9:Learning what you want to learn using programmable gradient information [PP/OL ].arXiv.(2024 -02-13) [2025 -02-12].https://arxiv.org/abs/ 2402.13616.