Truck License Plate Recognition Algorithm and Application Based on Large and Small Model Collaboration

Lin Bing; Yan Yan

doi:10.11959/j.issn.2096-0271.BDR25280

您当前的位置：

首页 >

文章列表页 >

Truck License Plate Recognition Algorithm and Application Based on Large and Small Model Collaboration

更新时间：2026-05-11

- Truck License Plate Recognition Algorithm and Application Based on Large and Small Model Collaboration
- Big Data Research (2026)
- 作者机构：
  
  联通（广东）产业互联网有限公司，广东广州 510000
- 作者简介：
- 基金信息：
- DOI：10.11959/j.issn.2096-0271.BDR25280
  CLC： U495
- Received：17 December 2025，
  
  Revised：2026-04-28，
  
  Accepted：11 May 2026，
- 稿件说明：
移动端阅览
Lin Bing, Yan Yan. Truck License Plate Recognition Algorithm and Application Based on Large and Small Model Collaboration[J/OL]. Big Data Research, 2026.
DOI：

Lin Bing, Yan Yan. Truck License Plate Recognition Algorithm and Application Based on Large and Small Model Collaboration[J/OL]. Big Data Research, 2026. DOI： 10.11959/j.issn.2096-0271.BDR25280.

摘要

针对公路大型货车治理中存在的多摄像头视角异构、目标检测任务复杂、传统算法泛化能力不足等问题，本文提出一种基于视觉语言大模型QWen-2.5-VL与YOLO11目标检测模型、PaddleOCRv5文字识别模型结合的多模型智能监控算法。该算法针对道路监控系统采集的车头、侧面、车尾多视角图像，实现吊车检测、车牌清晰度判定、车牌识别及车轴可见性判断四项核心功能。通过构建多任务独立并行推理框架，将视觉语言模型的语义理解能力与目标检测算法的精准定位能力分别应用于不同的任务场景，在真实场景数据集上的实验结果表明，算法在吊车检测任务中mAP达到94%，车牌清晰度判定F1值提升至96%，车轴可见性判断F1值达94.3%。本文的研究成果为公路货运车辆智能化监管提供了高效、可靠的技术方案，具有重要的工程应用价值。

Abstract

To address issues such as heterogeneous camera perspectives

complex object detection tasks

and insufficient generalization capabilities of traditional algorithms in the governance of large trucks on highways

this paper proposes a multi-model intelligent monitoring algorithm based on the deep integration of the visual language model QWen-2.5-VL

the YOLO11 object detection model

and the PaddleOCRv5 text recognition model. The algorithm targets multi-angle images collected by road monitoring systems

including front

side

and rear views

to achieve four core functions: crane detection

license plate clarity assessment

license plate recognition

and axle visibility judgment. By constructing a multi-task independent parallel processing framework

it applies the semantic understanding capability of the visual language model and the precise positioning ability of the object detection algorithm to different task scenarios respectively.Experimental results on real-world scenario datasets show that the algorithm achieves an mAP of 94% in crane detection

an F1 score improvement to 96% in license plate clarity assessment

and an F1 score of 94.3% in axle visibility judgment. The research outcomes provide an efficient and reliable technical solution for the intelligent supervision of highway freight vehicles

with significant engineering application value.

关键词

Keywords

references

高轶男 , 巩建强 . 我国道路交通事故特征及致因分析 [J ] . 安全与环境学报 , 2023 , 23 ( 11 ): 4013 - 4023 .

吴林 , 曹雯 . YOLO-SCDI:基于改进YOLOv8的车辆检测算法 [J ] . 电子与信息学报 , 2025 , 47 ( 10 ): 3967 - 3978 .

陈鑫鑫 , 郭晓雯 , 刘洪霞 . 基于深度学习的车辆检测算法研究 [J ] . 江苏通信 , 2024 , 40 ( 02 ): 114 - 118 .

许晨杰 , 金睿哲 , 韩帅峰 , 等 . 基于计算机视觉的钢铁行业车牌识别算法 [J ] . 信息技术与标准化 , 2022 ,( 11 ): 68 - 72 .

童宇行 . 基于边缘检测的车牌识别系统 [J ] . 信息记录材料 , 2025 , 26 ( 02 ): 184 - 186+207 .

庄凤云 . 基于视觉语言大模型的可进化GUI自动导航方法研究 [J ] . 江苏通信 , 2024 , 40 ( 05 ): 93 - 97 .

董禹彤 , 侯惠芳 , 龚明明 , 等 . 视觉语言大模型驱动的多模态工业缺陷检测与智能决策 [J ] . 计算机工程与应用 , 2026 , 62 ( 08 ): 176 - 188 .

Zhang S , Chi C , Yao Y , et al . Bridging the gap between anchor-based and anchor-free detection via adaptive training sample selection [C ] // Proceedings of the IEEE/CVF conference on computer vision and pattern recognition . 2020 : 9759 - 9768 .

Ma Y , Li Z , Zhang X Y , et al . PaddleOCR: A practical ultra-lightweight OCR system [C ] // Proceedings of the 28th ACM International Conference on Multimedia . New York : ACM , 2020 : 4181 – 4183 .

MEESAD P , THUMTHONG W . Advanced deep learning techniques for automated license plate recognition [J ] . Scientific Reports , 2025 , 15 ( 1 ): 24967 .

SATYA B , MANONGGA D , HENDRY , et al . Optimized YOLOv8 for Automatic License Plate Recognition on Resource Constrained Devices [J ] . Engineering , Technology & Applied Science Research, 2025 , 15 ( 2 ): 21976 - 21981 .

WANG X , WANG M , GUO H , et al . License plate recognition system for complex scenarios based on improved YOLOv5s and LPRNet [J ] . Scientific Reports , 2025 , 15 ( 1 ): 34741 .

MANE V , POKE A , RAIBAGKAR C , et al . Automatic number plate detection and recognition using YOLO world based convolutional neural networks [J ] . Computers and Electrical Engineering , 2024 , 120 ( Part A ): 109668 .

ZHANG X , YU S . A Lightweight License Plate Recognition Method Based on YOLOv8 [J ] . Electronics , 2025 , 14 ( 17 ): 3482 .

GANESH M A , SARAVANA PERUMAAL S , GOMATHI SANKAR S M . Automated license plate authentication framework using multi-view vehicle images [J ] . Journal of Intelligent & Fuzzy Systems , 2025 , 48 ( 5 ): 645 - 671 .

NASCIMENTO V , LIMA G E , RIBEIRO R O , et al . Toward advancing license plate super-resolution in real-world scenarios: a dataset and benchmark [J ] . Journal of the Brazilian Computer Society , 2025 , 31 ( 1 ): 435 - 449 .

TIJINK M , LEVENDEEV S , NIEUWENHUIS E , et al . Vision on the Move: Automated Hazardous Material Plate Detection in Freight Transport [C ] // Computer Analysis of Images and Patterns (CAIP 2025) . Cham : Springer Nature Switzerland , 2025 .

TIAN X Y , GU J R , LI B L , et al . DriveVLM: the convergence of autonomous driving and large vision-language models [C ] //AGRAWAL P, KROEMER O, BURGARD W (Eds.). Proceedings of The 8th Conference on Robot Learning. Proceedings of Machine Learning Research, Vol. 270 . [S.l.] : PMLR , 2025 : 4698 - 4726 .

GU Z P , ZHU B K , ZHU G B , et al . AnomalyGPT: detecting industrial anomalies using large vision-language models [C ] // Proceedings of the Thirty-Eighth AAAI Conference on Artificial Intelligence and Thirty-Sixth Conference on Innovative Applications of Artificial Intelligence and Fourteenth Symposium on Educational Advances in Artificial Intelligence . Palo Alto : AAAI Press , 2024 : 215 .

GUAN W , LAN J , CAO J , et al . EMIT: enhancing MLLMs for industrial anomaly detection via difficulty-aware GRPO [EB/OL ] . ( 2025-07-29 ) [ 2026-02-12 ] . https://arxiv.org/abs/2507.21619 https://arxiv.org/abs/2507.21619 .

张霞 , 乔环宇 , 曹峰 . 基于改进YOLOv8的高分辨率遥感图像目标检测算法 [J ] . 大数据 , 2025 , 11 ( 05 ): 152 - 169 .

彭森 , 张珂珂 , 杨一群等 . 基于改进YOLOv8的矿井复杂环境小目标检测算法 [J ] . 大数据 , 2025 , 11 ( 06 ): 166 - 181 .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

High resolution remote sensing image object detection algorithm based on improved YOLOv8

Related Author

QIAO Huanyu

CAO Feng

ZHANG Xia

Related Institution

School of Computer and Information Technology, Shanxi University

Postal code：100079
Tel：（010）53879208、53859533 Email：bdr@bjxintong.com.cn
Technical support is provided by Beijing Founder electronics co., LTD 京ICP备09064830号-19 京公网安备11010802024621
It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.
Total Visits：274499 Visits Today：88

⁰