|本期目录/Table of Contents|

[1]王明吉a,刘 博a,陈秋梦a,等.基于yolov3的车牌定位识别系统[J].工业仪表与自动化装置,2022,(01):97-100.[doi:10.19950/j.cnki.cn61-1121/th.2022.01.020]
　WANG Mingjia,LIU Boa,CHEN Qiumenga,et al.License plate location and recognition system based on yolov3[J].Industrial Instrumentation & Automation,2022,(01):97-100.[doi:10.19950/j.cnki.cn61-1121/th.2022.01.020]
点击复制

基于yolov3的车牌定位识别系统

《工业仪表与自动化装置》[ISSN:1000-0682/CN:61-1121/TH]

卷:
期数:
2022年01期
页码:
97-100
栏目:
出版日期:
2022-02-15

文章信息/Info

Title:
License plate location and recognition system based on yolov3
文章编号:
1000-0682（2022）01-0000-00
作者:
王明吉a; 刘 博a; 陈秋梦a; 王殿举b
东北石油大学 a.物理与电子工程学院；b.海洋油气工程学院，黑龙江 大庆 163318
Author(s):
WANG Mingjia; LIU Boa; CHEN Qiumenga; WANG Dianjub
?.School of Physics and Electronic Engineering,b.Sollege of Offshore Oil and Gas Engineering ,Northeastern Petroleum University, Heilongjiang Daqing 163318,China
关键词:
yolov3; 车牌定位; 模式识别; 车牌识别
Keywords:
yolov3;  license plate location;  pattern recognition;  license plate recognition
分类号:
TP399
DOI:
10.19950/j.cnki.cn61-1121/th.2022.01.020
文献标志码:
A
摘要:
为解决目前市场上车牌识别设备的高时延以及对高性能处理器高依赖性的问题，提出了一种基于yolov3的车牌定位识别方法，实现对车辆的车牌进行自动定位与识别。该方法通过收集5000张车辆图像数据，打包为VOC数据集，使用LabelImg工具对图像进行标注，并构建13个卷积层的yolo模型，在loss曲线趋于稳定之后，完成模型的训练，最终使用已训练的模型对车辆图像数据进行车牌识别。实验表明：在1000个测试数据中，该方法识别率的达到97.5%，平均耗时18.19 ms，能够快速精准的对车辆进行车牌识别。
Abstract:
In order to solve the problems of high time delay and high dependence on high-performance processor of license plate recognition equipment in the market, a license plate location and recognition method based on yolov3 is proposed to realize automatic location and recognition of vehicle license plate. In this method, 5000 vehicle image data are collected and packaged into VOC data set, the images are labeled with labelimg tool, and the Yolo model of 13 convolution layers is constructed. After the loss curve tends to be stable, the model training is completed, and finally the trained model is used to recognize the vehicle image data. The experimental results show that in 200 test data, the recognition rate of this method reaches 97.5%, and the average time is 17.19ms. It can recognize the vehicle license plate quickly and accurately.

参考文献/References:

[1]张松兰.车牌识别系统算法综述[J].电子技术与软件工程,2021(04):128-130.

[2]姚文凤,甄彤,吕宗旺,等.车牌字符分割与识别技术研究[J].现代电子技术,2020,43(19):65-69.
[3]李怡珍.车牌识别系统中定位方法的研究[J].电脑知识与技术,2020,16(26):178-179+188.
[4]贺智龙,肖中俊,严志国.轻量级车牌识别卷积网络[J].齐鲁工业大学学报,2020,34(04):35-41.
[5]董峻妃,郑伯川,杨泽静.基于卷积神经网络的车牌字符识别[J].计算机应用,2017,37(7):2014-2018.
[6]陈政.车牌识别系统研究综述[J].现代信息科技,2019,3(11):193-195.
[7]胡成伟,袁明辉.基于MSER与SVM算法的车牌定位识别方法[J].软件,2020,41(02):179-182.
[8]吕秀丽,杨士东.基于像素间双通道差异值的车牌定位算法[J].工业仪表与自动化装置,2020,(02):128-130.
[9]何旭光,罗一平,江磊.基于YOLO的前方车辆检测[J].舰船电子工程,2021,41(01):137-139.
[10]陈国栋,王翠瑜,张神德,等.基于改进YOLO V3的塔式起重机裂缝检测研究[J].贵州大学学报(自然科学版),2021,38(03):76-82.
[11]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.
[12]ROHAZIAT N , RAZALI M , WAN N , et al. White Blood Cells Detection using YOLOv3 with CNN Feature Extraction Models[J]. International Journal of Advanced Computer Science and Applications, 2020, 11(10).
[13]NAIK B T , HASHMI M F . Ball and Player Detection & Tracking in Soccer Videos Using Improved YOLOV3 Model. 2021.
[14]REDMON J, FARHADI A. YOLOv3: An Incremental Improvement[J]. arXiv e-prints, 2018.

相似文献/References:

[1]冯宇平,安雪美,孔祥茂.基于视频流的车牌识别研究及系统设计[J].工业仪表与自动化装置,2017,(01):32.
　FENG Yuping,AN Xuemei,KONG Xiangmao.Research on Vehicle License Plate Recognition and System Design Based on Video Stream[J].Industrial Instrumentation & Automation,2017,(01):32.
[2]吕秀丽,杨士东.基于像素间双通道差异值的车牌定位算法[J].工业仪表与自动化装置,2020,(02):128.
　Lü Xiuli,YANG Shidong.License plate location algorithm based on dual channel difference between pixels[J].Industrial Instrumentation & Automation,2020,(01):128.
[3]单文轩,徐永祥,白福忠,等.基于改进YOLOv3的工业指针式仪表检测[J].工业仪表与自动化装置,2023,(04):64.[doi:10.19950/j.cnki.cn61-1121/th.2023.04.012]
　SHAN Wenxuan,XU Yongxiang,BAI Fuzhong,et al.Industrial pointer meter detection based on improved YOLOv3[J].Industrial Instrumentation & Automation,2023,(01):64.[doi:10.19950/j.cnki.cn61-1121/th.2023.04.012]

备注/Memo

备注/Memo:
收稿日期：2021-08-13

作者简介：
王明吉(1963)，男，工学博士，现任东北石油大学物理与电子工程学院系主任，主演研究方向是光电检测及信息处理技术。
更新日期/Last Update: 1900-01-01