计算机科学

计算机科学 ›› 2024, Vol. 51 ›› Issue (6A): 220200162-7.doi: 10.11896/jsjkx.220200162

• 图像处理&多媒体技术 • 上一篇    下一篇

基于改进图像增强及CNN的复杂环境车牌识别算法

杨秀璋1, 武帅1, 任天舒1, 廖文婧1, 项美玉2, 于小民3, 刘建义1, 陈登建1   

  1. 1 贵州财经大学信息学院 贵阳 550025
    2 贵州财经大学大数据应用与经济学院(贵阳大数据金融学院) 贵阳 550025
    3 贵州财经大学贵州省经济系统仿真重点实验室 贵阳 550025
  • 发布日期:2024-06-06
  • 通讯作者: 武帅(472191973@qq.com)
  • 作者简介:(1455136241@qq.com)
  • 基金资助:
    贵州省科技计划项目(黔科合基础[2020]1Y279);贵州财经大学2021年度校级项目(2021KYQN03)

Complex Environment License Plate Recognition Algorithm Based on Improved Image Enhancement and CNN

YANG Xiuzhang1, WU Shuai1, REN Tianshu1, LIAO Wenjing1, XIANG Meiyu2, YU Xiaomin3, LIU Jianyi1, CHEN Dengjian1   

  1. 1 School of Information,Guizhou University of Finance and Economics, Guiyang 550025,China
    2 Guiyang School of Big Data and Finance,School of Big Data Application and Economics,Guizhou University of Finance and Economics, Guiyang 550025,China
    3 Guizhou Key Laboratory of Economics System Simulation,Guizhou University of Finance and Economics, Guiyang 550025,China
  • Published:2024-06-06
  • About author:YANG Xiuzhang,born in 1991,Ph.D.His main research interests include artificial intelligence,image identification and natural language processing.
    WU Shuai,born in 1994,Ph.D candidate.His main research interests include information service and computer application.
  • Supported by:
    Guizhou Science and Technology Plan Project(QiankeheFoundation[2020]1Y279) and Guizhou University of Finance and Economics Scientific Research Fund Project(2021KYQN03).

PDF (PC)

371

摘要： 传统图像识别和深度学习模型较难检测复杂环境下的车牌,其场景适用性和精确率较低,从而严重威胁交通安全,影响智慧交通发展。基于此,文中提出了一种基于改进图像增强及CNN的复杂环境车牌识别算法。首先,结合目标图像的平均灰度值,利用ACE算法和暗通道先验去雾算法对复杂环境下的车牌数据集进行图像增强;然后提出了一种融合色彩关键特征和波峰关键特征的车牌区域定位算法,通过8个核心步骤有效定位复杂环境下车牌的区域;最后构建五层卷积神经网络的模型,以实现车牌字符识别。实验结果表明,所提算法能有效识别复杂环境下行驶车辆的车牌,该算法在复杂环境车牌区域定位的精确率为86.04%,召回率为82.60%,F1值为84.29%,其F1值比传统图像处理算法提升了47.29%,比SSD算法提升了24.73%,比YOLO算法提升了26.37%,比YOLOv3提升了17.15%。同时,所提方法的时间复杂度较低,属于一种轻量级的车牌识别方法,能消除噪声并实现车牌字符识别,具有一定的应用前景和实用价值,也将为智慧交通的研究提供理论基础。

关键词: 车牌识别, 图像增强, 深度学习, 复杂环境, 智慧交通

Abstract: Traditional image recognition and deep learning models are difficult to detect license plates in complex environments.Their scene applicability and accuracy are low,which seriously threatens traffic safety and affects the development of intelligent transportation.This paper proposes a complex environment license plate recognition algorithm based on improved image enhancement and CNN.First,after calculating the average gray value of the target image,we use the ACE algorithm and the dark channel prior dehazing algorithm to perform image enhancement on the license plate dataset in complex environments.Then,a license plate area localization algorithm that combines the key features of color and the peak is proposed,effectively locating the license plate area by eight-core steps in a complex environment.Finally,a five-layer convolutional neural network model is constructed to recognize the license plate character.Experimental results show that the proposed algorithm can effectively identify the license plates of vehicles in complex environments.The precision of the algorithm’s license plate area location in complex environments is 86.04%,the recall is 82.60%,and the F1-score is 84.29%.Among them,the F1-score of the proposed algorithm is 47.29% higher than the traditional image processing algorithm,24.73% higher than the SSD algorithm,26.37% higher than the YOLO algorithm and 17.15% higher than the YOLOv3 algorithm.At the same time,the time complexity of the proposed method is low,and it belongs to a lightweight license plate recognition method.Also,it can eliminate noise and realize license plate character re-cognition.Therefore,it has specific application prospects and practical value and provides a theoretical basis for intelligent transportation research.

Key words: License plate recognition, Image enhancement, Deep learning, Complex environment, Intelligent transportation

中图分类号: 

  • TP391
[1]ZHANG B,PANG J M,ZHANG W S,et al.Application of big data technology in the development of intelligent transportation[J].Science & Technology Review,2020,38(9):47-54.
[2]CHEN J Y,SHEN S J,SU M M,et al.Black-box AdversarialAttack on License Plate Recognition System[J].Acta Automatica Sinica,2021,47(1):121-135.
[3]PEI Y L,ZHANG T.Vehicle Numb:er Plate and Vehicle Identification System based on Artificial Intelligence[J].Highway,2019,64(8):277-281.
[4]BAI R,XU Y C,LI Y L,et al.Digital Character Recognition Technique for Intelligent Vehicles in Road Scenes[J].Laser & Optoelectronics Progress,2019,56(19):211-219.
[5]ZHANG Y,LUO X Y,WANG J W,et al.Research progress on digital image robust steganography[J].Journal of Image and Graphics,2022,27(1):3-26.
[6]WANG K,YANG F,JIANG S.Review of optical character recognition[J].Application Research of Computers,2020,37(S2):22-24.
[7]WANG X Y,LIU S J.Simulation of Super-Resolution ImageNoise Recognition Based on Multi-Directional Threshol[J].Computer Simulation,2021,38(12):132-135,181.
[8]ZHOU Q ,ZHOU H,XU L J,et al.Method for image Locating based on SURF algorithm and SLSH algorithm[J].Instrument Technique and Sensor,2020(10):103-107.
[9]XU J D,ZHAO T Y,FENG G Z,et al.Image Segmentation Algorithm Based on Context Fuzzy C-Means Clustering[J].Journal of Electronics & Information Technology,2021,43(7):2079-2086.
[10]GE Y Z,LIU H,WANG Y,et al.Survey on Deep Learning Image Recognition in Dilemma of Small Samples[J].Journal of Software,2022,33(1):193-210.
[11]LI N,ZHOU L H,YANG G,et al.An Improved YOLOv3 Anomaly Detection Method for Deck Lights[J].Ship Engineering,2021,43(S2):52-58,61.
[12]YANG X Z,XIA H,YU X M.Image Enhancement and Recognition Method Based on Shui-characters[J].Computer Science,2019,46(S2):324-328.
[13]GIRSHICK R.Fast r-cnn[C]//Proceedings of the IEEE international conference on computer vision.2015:1440-1448.
[14]BRUNELLI R.Template matching techniques in computer vision:theory and practice[M].John Wiley & Sons,2009.
[15]TYLER L K,CHIU S,ZHUANG J,et al.Objects and categories:feature statistics and object processing in the ventral stream[J].Journal of Cognitive Neuroscience,2013,25(10):1723-1735.
[16]ALPAYDIN E.Machine learning[M].MIT Press,2021.
[17]ALBAWI S,MOHAMMED T A,AL-ZAWI S.Understandingof a convolutional neural network[C]//2017 International Conference on Engineering Tnd technology(ICET).IEEE,2017:1-6.
[18]FU J Q,FENG R.Segmenting characters on license plate with low image quality based on conditional random fields[J].Computer Applications and Software,2014,31(10):157-160.
[19]XU J M,WEI X,LIN Y J,et al.Urban Vehicle Trip Chain Reconstruction based on Gradient Boosting Decision Tree[J].Journal of South China University of Technology(Natural Science Edition),2020,48(7):55-64.
[20]LI M K,KE Z X,YU D R,et al.Vehicle license plate number recognition algorithm based on mobile edge calculation[J].Computer Engineering and Design,2021,42(11):3151-3157.
[21]SU Y T,YU H Y,CHI J Z.Reseach on license plate recognition system based on LSTM algorithm[J].Electronic Measurement Technology,2021,44(18):67-71.
[22]ZHANG C Z,LI Y,KANG B L,et al.Blurred license PlateCharacter Recognition algorithm based on deep learning[J].Laser & Optoelectronics Progress,2021,58(16):259-266.
[23]LI Y,WANG X L,LEI L,et al.Research of license plate recognition method based on polarity detection and improved CNN framework[J].Electronic Measurement Technology,2021,44(14):26-32.
[24]REDMON J,FARHADI A.Yolov3:An incremental improve-ment[J].arXiv:1804.02767,2018.
[25]RAO W J,GU Y H,ZHU T T,et al.Intelligent License Plate Recognition Method in Complex Environment[J].Journal of Chongqing University of Technology(Natural Science),2021,35(3):119-127.
[26]RIZZI A,GATTA C,MARINI D.A new algorithm for unsupervised global and local colorcor-rection[C]//Pattern Recognition Letters.2003:1663-1677.
[27]HE K M,SUN J,TANGX O.Single image haze removal using dark channel prior[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2011,33(12):2341-2353.
[28]HE T,ZHAO T,XU H.Novel Algorithm of Single Image Dehazing Based on Dark Channel Prior[J].Computer Science,2021,48(7):219-224.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发