基于改进YOLOv3-SPP算法的道路车辆检测

TN92; 针对在城市道路场景下视觉检测车辆时,车辆密集和远处车辆呈现小尺度,导致出现检测精度低或者漏检的问题,提出了一种基于改进的YOLOv3-SPP算法,对激活函数进行优化,以DIOU-NMS Loss作为边界框损失函数,增强网络的表达能力.为提高所提算法对小目标和遮挡目标的特征提取能力,引入空洞卷积模块,增大目标的感受野.实验结果表明,所提算法在检测车辆目标时 mAP 提高了 1.79%,也有效减少了在检测紧密车辆目标时出现的漏检现象....

Full description

Saved in:
Bibliographic Details
Published in通信学报 Vol. 45; no. 2; pp. 68 - 78
Main Authors 王涛, 冯浩, 秘蓉新, 李林, 何振学, 傅奕茗, 吴姝
Format Journal Article
LanguageChinese
Published 北京信息科技大学信息与通信工程学院,北京 100101 25.02.2024
北京信息科技大学光电测试技术及仪器教育部重点实验室,北京 100192%国家计算机网络应急技术处理协调中心,北京 100029%河北农业大学河北省农业大数据重点实验室,河北 保定 071001
Subjects
Online AccessGet full text
ISSN1000-436X
DOI10.11959/j.issn.1000-436x.2024046

Cover

Abstract TN92; 针对在城市道路场景下视觉检测车辆时,车辆密集和远处车辆呈现小尺度,导致出现检测精度低或者漏检的问题,提出了一种基于改进的YOLOv3-SPP算法,对激活函数进行优化,以DIOU-NMS Loss作为边界框损失函数,增强网络的表达能力.为提高所提算法对小目标和遮挡目标的特征提取能力,引入空洞卷积模块,增大目标的感受野.实验结果表明,所提算法在检测车辆目标时 mAP 提高了 1.79%,也有效减少了在检测紧密车辆目标时出现的漏检现象.
AbstractList TN92; 针对在城市道路场景下视觉检测车辆时,车辆密集和远处车辆呈现小尺度,导致出现检测精度低或者漏检的问题,提出了一种基于改进的YOLOv3-SPP算法,对激活函数进行优化,以DIOU-NMS Loss作为边界框损失函数,增强网络的表达能力.为提高所提算法对小目标和遮挡目标的特征提取能力,引入空洞卷积模块,增大目标的感受野.实验结果表明,所提算法在检测车辆目标时 mAP 提高了 1.79%,也有效减少了在检测紧密车辆目标时出现的漏检现象.
Abstract_FL Aiming at the problem of low detection accuracy or missing detection caused by dense vehicles and small scale of distant vehicles in the visual detection of urban road scenes,an improved YOLOv3-SPP algorithm was proposed to op-timize the activation function and take DIOU-NMS Loss as the boundary frame loss function to enhance the expression ability of the network.In order to improve the feature extraction ability of the proposed algorithm for small targets and occluding targets,the void convolution module was introduced to increase the receptive field of the target.Based on the experimental results,the proposed algorithm improves the mAP by 1.79%when detecting vehicle targets,and also effec-tively reduce the missing phenomenon when detecting tight vehicle targets.
Author 冯浩
吴姝
李林
秘蓉新
何振学
王涛
傅奕茗
AuthorAffiliation 北京信息科技大学信息与通信工程学院,北京 100101;北京信息科技大学光电测试技术及仪器教育部重点实验室,北京 100192%国家计算机网络应急技术处理协调中心,北京 100029%河北农业大学河北省农业大数据重点实验室,河北 保定 071001
AuthorAffiliation_xml – name: 北京信息科技大学信息与通信工程学院,北京 100101;北京信息科技大学光电测试技术及仪器教育部重点实验室,北京 100192%国家计算机网络应急技术处理协调中心,北京 100029%河北农业大学河北省农业大数据重点实验室,河北 保定 071001
Author_FL MI Rongxin
HE Zhenxue
WANG Tao
LI Lin
FENG Hao
FU Yiming
WU Shu
Author_FL_xml – sequence: 1
  fullname: WANG Tao
– sequence: 2
  fullname: FENG Hao
– sequence: 3
  fullname: MI Rongxin
– sequence: 4
  fullname: LI Lin
– sequence: 5
  fullname: HE Zhenxue
– sequence: 6
  fullname: FU Yiming
– sequence: 7
  fullname: WU Shu
Author_xml – sequence: 1
  fullname: 王涛
– sequence: 2
  fullname: 冯浩
– sequence: 3
  fullname: 秘蓉新
– sequence: 4
  fullname: 李林
– sequence: 5
  fullname: 何振学
– sequence: 6
  fullname: 傅奕茗
– sequence: 7
  fullname: 吴姝
BookMark eNrjYmDJy89LZWBQNDTQMzS0NLXUz9LLLC7O0zM0MDDQNTE2q9AzMjAyMTAxY2HghIlFcDDwFhdnJhmYGhqbmxkYG3IyGD6dv-vJrr5nU3a-2D870t_Hv8xYNzgg4Pm66c82T30-q-Vl4-QX29e_2Lvsxb62Z4sbnm3t5mFgTUvMKU7lhdLcDKFuriHOHro-_u6ezo4-usVA68x0zS0tzA3SjM0NUtOSzJLSkkxTk0wtkg2SLZNNUkwMkiySjU1Mkk2N0yyB0oZpiSkmSSnGhpbGlikWyZZpJhamxtwMqhBzyxPz0hLz0uOz8kuL8oA2xpdUVCSBPWcEtMgYAIwjV2M
ClassificationCodes TN92
ContentType Journal Article
Copyright Copyright © Wanfang Data Co. Ltd. All Rights Reserved.
Copyright_xml – notice: Copyright © Wanfang Data Co. Ltd. All Rights Reserved.
DBID 2B.
4A8
92I
93N
PSX
TCJ
DOI 10.11959/j.issn.1000-436x.2024046
DatabaseName Wanfang Data Journals - Hong Kong
WANFANG Data Centre
Wanfang Data Journals
万方数据期刊 - 香港版
China Online Journals (COJ)
China Online Journals (COJ)
DatabaseTitleList
DeliveryMethod fulltext_linktorsrc
DocumentTitle_FL Road vehicle detection based on improved YOLOv3-SPP algorithm
EndPage 78
ExternalDocumentID txxb202402006
GroupedDBID -0Y
2B.
4A8
92I
93N
ALMA_UNASSIGNED_HOLDINGS
CCEZO
CUBFJ
GROUPED_DOAJ
PSX
TCJ
ID FETCH-LOGICAL-s1006-79870f370efb6bfb5eb58c0c9c4d40b8c344c53f9fb61fad4bd31939d8c9f4853
ISSN 1000-436X
IngestDate Thu May 29 04:00:49 EDT 2025
IsPeerReviewed false
IsScholarly true
Issue 2
Keywords YOLOv3-SPP算法
vehicle detection
atrous convolution
activation function
deep learning
深度学习
车辆检测
空洞卷积
YOLOv3-SPP algorithm
激活函数
Language Chinese
LinkModel OpenURL
MergedId FETCHMERGED-LOGICAL-s1006-79870f370efb6bfb5eb58c0c9c4d40b8c344c53f9fb61fad4bd31939d8c9f4853
PageCount 11
ParticipantIDs wanfang_journals_txxb202402006
PublicationCentury 2000
PublicationDate 2024-02-25
PublicationDateYYYYMMDD 2024-02-25
PublicationDate_xml – month: 02
  year: 2024
  text: 2024-02-25
  day: 25
PublicationDecade 2020
PublicationTitle 通信学报
PublicationTitle_FL Journal on Communications
PublicationYear 2024
Publisher 北京信息科技大学信息与通信工程学院,北京 100101
北京信息科技大学光电测试技术及仪器教育部重点实验室,北京 100192%国家计算机网络应急技术处理协调中心,北京 100029%河北农业大学河北省农业大数据重点实验室,河北 保定 071001
Publisher_xml – name: 北京信息科技大学光电测试技术及仪器教育部重点实验室,北京 100192%国家计算机网络应急技术处理协调中心,北京 100029%河北农业大学河北省农业大数据重点实验室,河北 保定 071001
– name: 北京信息科技大学信息与通信工程学院,北京 100101
SSID ssib051376031
ssj0002912165
ssib058759023
ssib001102965
ssib023646527
ssib023168036
ssib036439991
ssib050281523
ssib000968473
Score 2.4243064
Snippet TN92; 针对在城市道路场景下视觉检测车辆时,车辆密集和远处车辆呈现小尺度,导致出现检测精度低或者漏检的问题,提出了一种基于改进的YOLOv3-SPP算法,对激活函数进行优化,...
SourceID wanfang
SourceType Aggregation Database
StartPage 68
Title 基于改进YOLOv3-SPP算法的道路车辆检测
URI https://d.wanfangdata.com.cn/periodical/txxb202402006
Volume 45
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVAON
  databaseName: DOAJ Directory of Open Access Journals
  issn: 1000-436X
  databaseCode: DOA
  dateStart: 20230101
  customDbUrl:
  isFulltext: true
  dateEnd: 99991231
  titleUrlDefault: https://www.doaj.org/
  omitProxy: true
  ssIdentifier: ssj0002912165
  providerName: Directory of Open Access Journals
– providerCode: PRVHPJ
  databaseName: ROAD: Directory of Open Access Scholarly Resources
  issn: 1000-436X
  databaseCode: M~E
  dateStart: 19800101
  customDbUrl:
  isFulltext: true
  dateEnd: 99991231
  titleUrlDefault: https://road.issn.org
  omitProxy: true
  ssIdentifier: ssib058759023
  providerName: ISSN International Centre
link http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnV3Na9RAFA-1gngRRcXPUsE5SeokM5PMO07aLEXUFmyhglB2kk09rWC3UnpSEC-CFxVBDx4Ej3oQEQvqP6Pb9b_wvUk2ibbixyXMzseb95HZ95vJzDzPO19wG8fW9nzBc-lLy4VvdRT4XcszK2ORc3epz5Wr0fyyvLSiVib23WjtWtoY2Jlsa89zJf9jVcxDu9Ip2X-wbE0UMzCN9sUnWhiff2VjlioGHZYYlkp66pSlEQNMA0s1SzoMkusLlxfuCP_a4iJLY2ZSBjFVSgQDRTmAzSRLASElA-GaYbWOS8wxE7lEynRErYygnRHUXDGdtIGto8CJGrHSYSYg5kxJIWLaMKPGxqVukVWdOErIcNKUKOqJund9GGi3McitJn6QTw1OVKzFmyqYMTdWAgnaXtIIpTsiruqX0PU169ThlGdmW6xHpA5io-w1qGQg4ZErSZmNeLtbYY52jOylFNKvU73RTgc1HWAAzIT41u_m7ULg7ulr-RA6rC-FC9BYO5nyzsxqMIUtj1EGFaqwRxnNaLdXAwXOrRH9mTH9zRnSHZe_3CTusMlgc9O60vI--v0h-jzeWnCoJq8ITdqnqnkIzTfvkEKaIb5pfkcyUmENVgVhWWhuWlKIVBEM1vRUQPutmm_MCmfGUEWocjgphCAMXKzXWmMHvHNjeS_-Tlp3uq5fdPtrLSC4dNg7VM3gpk05HI94E1s3j3rBt5fbX7cfDZ98HH150Qy5nTfPhu-e7jy___3e49GHt6NPr0efHwxf3R2-f3jMW-6kS7PzfhWNxF8PaNktBnRthYh5r7CRLazqWaUznkEmc8mtzoSUmRIFYHFQdHNpc3RvAnKdQSERFR_3Jvu3-r0T3nQUR0pp6PIYpyNWqW6kbdFTXauyLBAyP-lNVSKuVv8266s_WfTUnyqc9g42A-uMNzm4vdE7i-h5YKfcS_ADdXiKRg
linkProvider Directory of Open Access Journals
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=%E5%9F%BA%E4%BA%8E%E6%94%B9%E8%BF%9BYOLOv3-SPP%E7%AE%97%E6%B3%95%E7%9A%84%E9%81%93%E8%B7%AF%E8%BD%A6%E8%BE%86%E6%A3%80%E6%B5%8B&rft.jtitle=%E9%80%9A%E4%BF%A1%E5%AD%A6%E6%8A%A5&rft.au=%E7%8E%8B%E6%B6%9B&rft.au=%E5%86%AF%E6%B5%A9&rft.au=%E7%A7%98%E8%93%89%E6%96%B0&rft.au=%E6%9D%8E%E6%9E%97&rft.date=2024-02-25&rft.pub=%E5%8C%97%E4%BA%AC%E4%BF%A1%E6%81%AF%E7%A7%91%E6%8A%80%E5%A4%A7%E5%AD%A6%E4%BF%A1%E6%81%AF%E4%B8%8E%E9%80%9A%E4%BF%A1%E5%B7%A5%E7%A8%8B%E5%AD%A6%E9%99%A2%2C%E5%8C%97%E4%BA%AC+100101&rft.issn=1000-436X&rft.volume=45&rft.issue=2&rft.spage=68&rft.epage=78&rft_id=info:doi/10.11959%2Fj.issn.1000-436x.2024046&rft.externalDocID=txxb202402006
thumbnail_s http://utb.summon.serialssolutions.com/2.0.0/image/custom?url=http%3A%2F%2Fwww.wanfangdata.com.cn%2Fimages%2FPeriodicalImages%2Ftxxb%2Ftxxb.jpg