计算机科学

计算机科学 ›› 2023, Vol. 50 ›› Issue (11A): 220700147-8.doi: 10.11896/jsjkx.220700147

• 图像处理&多媒体技术 • 上一篇    下一篇

复杂环境下自适应去雾的YOLOv3汽车识别算法

杨秀璋1, 武帅2, 李娜3, 杨雯雯4, 廖文婧1, 周继松1   

  1. 1 贵州财经大学信息学院 贵阳 550025
    2 南京 农业大学信息管理学院 南京 210003
    3 中国船舶工业系统工程研究院 北京 100094
    4 陕西省图书馆 西安 710000
  • 发布日期:2023-11-09
  • 通讯作者: 武帅(472191973@qq.com)
  • 作者简介:(1455136241@qq.com)
  • 基金资助:
    贵州省省级科技计划项目(黔科合基础[2020]1Y279);贵州财经大学2021年度校级项目(2021KYQN03)

YOLOv3 Vehicle Recognition Algorithm for Adaptive Dehazing in Complex Environments

YANG Xiuzhang1, WU Shuai2, LI Na3, YANG Wenwen4, LIAO Wenjing1, ZHOU Jisong1   

  1. 1 School of Information,Guizhou University of Finance and Economics,Guiyang 550025,China
    2 School of Information Management,Nanjing Agricultural University,Nanjing 210003,China
    3 Systems Engineering Research Institute,Beijing 100094,China
    4 Shaanxi Library,Xi'an 710000,China
  • Published:2023-11-09
  • 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(qiankehe foundation [2020]1Y279) and Guizhou University of Finance and Economics Scientific Research Fund Project(2021KYQN03).

PDF (PC)

215

摘要： 复杂环境因素会严重影响道路汽车目标检测算法的性能,传统方法识别精度较低且感知慢,严重威胁交通安全,为此提出一种融合自适应图像去雾的YOLOv3汽车识别算法。首先,在图像预处理阶段构建自适应图像去雾算法,融合ACE去雾算法和暗通道去雾算法,有效降低雨雾图像噪声;其次,利用改进的YOLOv3算法识别和定位汽车位置;最后,通过详细的对比实验证明方法的有效性,并准确识别出复杂天气中行驶的车辆。实验结果表明,所提方法能有效降低雨雾情况下的噪声,对行驶车辆进行有效定位,其精确率、召回率和F1值分别为0.944,0.934和0.939,均高于传统SSD,YOLO和YOLOv3算法,并且具有较好的鲁棒性和速度,这将为智慧交通的发展提供理论基础并具有实践意义。

关键词: 深度学习, 自适应图像去雾, YOLOv3, 汽车识别, 目标检测

Abstract: In view of the complex environmental factors will seriously affect the performance of road vehicle target detection algorithms,traditional methods have low recognition accuracy and slow perception,which seriously threatens traffic safety.This paper proposes a YOLOv3 vehicle recognition algorithm based on adaptive image dehazing.First,an adaptive image dehazing algorithm is constructed in the image preprocessing stage.The ACE dehazing algorithm and the dark channel dehazing algorithm are combined to effectively reduce the noise of rain and fog images.Second,the improved YOLOv3 algorithm is used to identify and locate the car position.Finally,the effectiveness of the method is demonstrated through detailed comparative experiments,and the vehicles driving in complex weather are accurately identified.Experimental results show that the proposed method can effectively reduce the noise in rain and fog conditions,and can effectively locate the driving vehicle.Its precision,recall and F1 value is 0.944,0.934 and 0.939,respectively,which is higher than that of traditional SSD,YOLO and YOLOv3 algorithms.It has good robustness and speed,which will provide theoretical basis for the development of intelligent transportation and has practical significance.

Key words: Deep learning, Adaptive image dehazing, YOLOv3, Vehicle recognition, Target detection

中图分类号: 

  • TP391.4
[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]MA C X,WANG Y J,WANG X.Identification of Coal Vehicles Based on Convolutional Neural Network[J].Computer Science,2020,47(S2):219-223.
[3]ZHANG X R,CHEN X,SUN W,et al.Progress of Vehicle Re-identification Research Based on Deep Learning[J].Computer Engineering,2020,46(11):1-11.
[4]CHENG Q,FAN Y,LIU Y C,et al.Multi-feature fusion vehicle identification technology[J].Infrared and Laser Engineering,2018,47(7):316-321.
[5]LAN H,FANG Z Y.Recent Advances in Zero-Shot Learning[J].Journal of Electronics & Information Technology,2020,42(5):1188-1200.
[6]YUAN G L,HOU J,YIN K Y.Neight-Time Aerial Image Vehicle Recognition Technology Based on Transfer Learning and Image Enhancement[J].Journal of Computer-Aided Design & Computer Graphics,2019,31(3):467-473.
[7]LI Y,XU Q K,LI K D.New method of Residual Dense Generative Adversarial Networks for Image Restoration[J].Journal of Chinese Computer Systems,2020,41(4):830-836.
[8]LI M X,LIN Z K,QU Y.Survey of Vehicle Object Detection Algorithm in Computer Vision[J].Computer Engineering and Applications,2019,55(24):20-28.
[9]BI X L,QIU Y L,XIAO B,et al.Histogram Equalization Detection Based on Statistical Features in Digital Image[J].Chinese Journal of Computers,2021,44(2):292-303.
[10]YANG Y,ZHANG B S,ZHOU J,et al.An optical compensation based quick dehazing algorithm using channel prior[J].Computer Engineering & Science,2018,40(11):2033-2039.
[11]LIU Z X,WANG Z H.Machine Learning Feedback Based Vehicle Automated Road Conditions Recognition[J].Computer Simulation,2012,29(1):339-343.
[12]CHEN Z Z,ZHANG S G,DU F,et al.Monitoring System of Freeway Vehicle Lane Separation Based on Video Image Detection[J].Science Technology and Engineering,2021,21(9):3682-3688.
[13]YANG W,GONG J Q,WEI L.Preceding vehicle image Recognition Based on multivariate feature information fusion[J].Journal of Chang'an University(Natural Science Edition),2016,36(04):79-85.
[14]CHEN J Q,JIN X H,WANG W Y,et al.Vehicle Flow Detection Based on YOLOv3 and DeepSort[J].Acta Metrologica Sinica,2021,42(6):718-723.
[15]HUANG Y Z,WANG N Z,LIANG T C,et al.Vehicle Recognition Method Based on Improved CenterNet[J].Journal of South China University of Technology(Natural Science Edition),2021,49(7):94-102.
[16]MAO Q C,JIA R S,ZUO X Q,et al.A Traffuc Syrveillance Video Vehicle Detection Method Based on Deep Learning[J].Computer Applications and Software,2020,37(9):111-117,164.
[17]WANG K Y,HAN X H.Real-time Detection Based on Optimized You Only Look Once v2 Algorithm[J].Journal of University of Jinan(Science and Technology),2020,34(5):443-449.
[18]ZHUANG X L,TAN F K,LI Z,et al.Image Defogging Algorithm Based on Dark Channel Prior and Optimized Auto-color[J].Computer Applications and Software,2021,38(7):190-195.
[19]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.
[20]YANG X Z,WU S,XIA H,et al.Research on Shui Characters Extraction and Recognition Based on Adaptive Image Enhancement Technology[J].Computer Science,2021,48(S1):74-79.
[21]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.
[22]GAO Z Z,WEI W B,PAN Z K,et al.Image dehazing combining dark channel prior and Hessian regular term[J].Journal of Graphics,2020,41(1):73-80.
[23]XIE L,XIONG G,YU B,et al.A Novel Haze Removal Algorithm for Atmospheric Degraded Image with Dark Channel Prior[J].Control Engineering of China,2020,27(2):207-211.
[24]LIU W J,BAI W S,QU H C,et al.Image dehazing based on GF-MSRCR and dark channel prior[J].Journal of Image and Graphics,2019,24(11):1893-1905.
[25]LI S S,LIU H R,GAN Y D,et al.Image Dehazing Network Based on Residual Dense Block and Attention Mechanism[J].Journal of Hunan University(Natural Sciences),2021,48(6):112-118.
[26]WANG B Z,SHI L Y,GUO Z T,et al.Application of improved Faster R-CNN in vehicle recognition[J].Modern Electronics Technique,2019,42(23):48-52.
[27]CHEN Z K.Homomorphic Filtering for Navigation-Mark Image Dehazing with Convolutional Neural Network[J].Navigation of China,2020,43(4):84-88.
[28]CHEN Q J,ZHANG X.Single Image Dehazing Based on Multiple Convolutional Neural Networks[J].Acta Automatica Sinica,2021,47(7):1739-1748.
[29]CHEN M,CHEN X,YU H M.Application of SSD Network Algorithm Model in Vehicle Axle Type Recognition[J].Journal of Ordnance Equipment Engineering,2021,42(8):227-232.
[30]LAN W,DANG J,WANG Y,et al.Pedestrian detection based on YOLO network model[C]//2018 IEEE International Conference on Mechatronics and Automation (ICMA).IEEE,2018:1547-1551.
[31]ZHOU X,LIU S D,PAN W,et al.Vehicle Color Recognition in Natural Traffic Scene[J].Computer Science,2021,48(S1):15-20,37.
[32]QIU M K,LI X Y.Detail-aware discriminative feature learning model for vehicle re-identification[J].Acta Scientiarum Naturalium Universitatis Sunyatseni,2021,60(4):111-120.
[33]NING J,WANG N,ZHU M.Vehicle type recognition algorithm based on improved Faster R-CNN[J].Journal of Anhui University(Natural Science Edition),2021,45(3):26-33.
[34]TANG X Y,SONG H S,WANG W,et al.3D Vehicle Information Recognition Algorithm of Monocular Camera Based on Self-Calibration in Traffic Scene[J].Journal of Computer-Aided Design & Computer Graphics,2020,32(8):1305-1314.
[35]RIZZI A,GATTA C,MARINI D.A new algorithm for unsupervised global and local color cor-rection[C].Pattern Recognition Letters,2003(124):1663-1677.
[36]GETREUER P.Automatic Color Enhancement (ACE) and its Fast Implementation[C].Image Process,2012(2):266-277.
[37]YIN S N,CUI X R,LI J,et al.Image enhancement method of visual odometer based on fast ACE algorithm[J].Journal of Electronic Measurement and Instrumentation,2021,35(6):27-33.
[38]He K M,SUN J,TANG X O.Single image haze removal using dark channel prior[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2011,33(12):2341-2353.
[39]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.
[40]HUANG G,LIU Z,WEINBERGER K Q,et al.Densely connected convolutional networks[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition,2017.
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