融合多尺度特征的YOLOv8裂缝缺陷检测算法

TP391.4; 针对井壁裂缝背景复杂,纵横比差异大,导致检测效率低、漏检等问题,提出了一种融合多尺度特征的裂缝缺陷检测模型EDG-YOLO.设计特征提取模块EIRBlock(efficient inverted residual block),并构建C2fEIR增强主干网络对井壁浅层裂缝特征信息的提取能力.在颈部融合CSP_EDRAN(CSP efficient dilated reparam aggregation network)实现对裂缝特征信息的复用,促进浅层与深层语义信息之间的交互.嵌入DAM(dual attention module)注意力机制,增强井壁裂缝特征的表达能力.构建...

Full description

Saved in:
Bibliographic Details
Published in计算机工程与应用 Vol. 60; no. 22; pp. 261 - 270
Main Authors 赵佰亭, 程瑞丰, 贾晓芬
Format Journal Article
LanguageChinese
Published 安徽理工大学 电气与信息工程学院,安徽 淮南 232001%安徽理工大学 人工智能学院,安徽 淮南 232001 15.11.2024
Subjects
井壁裂缝
注意力机制
shaft lining crack
attention mechanism
目标检测
YOLOv8
object detection
轻量化
lightweight
Online AccessGet full text
ISSN1002-8331
DOI10.3778/j.issn.1002-8331.2404-0332

Cover

Abstract TP391.4; 针对井壁裂缝背景复杂,纵横比差异大,导致检测效率低、漏检等问题,提出了一种融合多尺度特征的裂缝缺陷检测模型EDG-YOLO.设计特征提取模块EIRBlock(efficient inverted residual block),并构建C2fEIR增强主干网络对井壁浅层裂缝特征信息的提取能力.在颈部融合CSP_EDRAN(CSP efficient dilated reparam aggregation network)实现对裂缝特征信息的复用,促进浅层与深层语义信息之间的交互.嵌入DAM(dual attention module)注意力机制,增强井壁裂缝特征的表达能力.构建轻量级检测头GDetect,借助GSConv模块进一步轻量化网络.在自制井壁裂缝数据集上的实验结果表明,与YOLOv8相比,EDG-YOLO的平均检测精度达到87.4％,提高了 2.3个百分点,模型的参数量和计算量分别降低了 33％和47％,单幅图像推理时间为13.2 ms,满足井下场景的实时检测需求.
AbstractList TP391.4; 针对井壁裂缝背景复杂,纵横比差异大,导致检测效率低、漏检等问题,提出了一种融合多尺度特征的裂缝缺陷检测模型EDG-YOLO.设计特征提取模块EIRBlock(efficient inverted residual block),并构建C2fEIR增强主干网络对井壁浅层裂缝特征信息的提取能力.在颈部融合CSP_EDRAN(CSP efficient dilated reparam aggregation network)实现对裂缝特征信息的复用,促进浅层与深层语义信息之间的交互.嵌入DAM(dual attention module)注意力机制,增强井壁裂缝特征的表达能力.构建轻量级检测头GDetect,借助GSConv模块进一步轻量化网络.在自制井壁裂缝数据集上的实验结果表明,与YOLOv8相比,EDG-YOLO的平均检测精度达到87.4％,提高了 2.3个百分点,模型的参数量和计算量分别降低了 33％和47％,单幅图像推理时间为13.2 ms,满足井下场景的实时检测需求.
Abstract_FL To solve the problems of low detection efficiency and missing detection caused by complex background and large aspect ratio difference of shaft lining cracks,a crack defect detection model EDG-YOLO with multi-scale features is proposed.Firstly,the feature extraction module EIRBlock(efficient inverted residual block)is designed,and C2fEIR is constructed to enhance the ability of backbone network to extract the shallow crack feature information.Secondly,the CSP_EDRAN(CSP efficient dilated reparam aggregation network)is fused in the neck to realize the reuse of the crack feature information,and promote the interaction between the shallow and deep semantic information.Meanwhile,the attention mechanism of DAM(dual attention module)is embedded to enhance the expression ability of shaft lining crack features.Finally,a lightweight detection head GDetect is constructed,and the network is further lightweight with the help of GSConv module.The experimental results on the self-made shaft lining crack dataset show that,compared with YOLOv8,the average detection accuracy of EDG-YOLO is 87.4％,which is increased by 2.3 percentage points,the num-ber of parameters and the amount of calculation of the model are reduced by 33％and 47％respectively.The inference time of a single image is 13.2 ms,which meets the real-time detection requirements of downhole scenes.
Author 贾晓芬
程瑞丰
赵佰亭
AuthorAffiliation 安徽理工大学 电气与信息工程学院,安徽 淮南 232001%安徽理工大学 人工智能学院,安徽 淮南 232001
AuthorAffiliation_xml – name: 安徽理工大学 电气与信息工程学院,安徽 淮南 232001%安徽理工大学 人工智能学院,安徽 淮南 232001
Author_FL CHENG Ruifeng
JIA Xiaofen
ZHAO Baiting
Author_FL_xml – sequence: 1
  fullname: ZHAO Baiting
– sequence: 2
  fullname: CHENG Ruifeng
– sequence: 3
  fullname: JIA Xiaofen
Author_xml – sequence: 1
  fullname: 赵佰亭
– sequence: 2
  fullname: 程瑞丰
– sequence: 3
  fullname: 贾晓芬
BookMark eNo9jT1Lw0Ach2-oYK39Em4OiXf3v_QuoxTfIJBFB6eS3EtpkCt4qGRTcVBE6CJiQZeKowgiYih-myT1W1hQXH4PPMPzW0INO7QaoRWCfeBcrGX-wDnrE4ypJwCITxlmHgagDdT8t4uo7dwgxQEBHnAIm6gze7wpR5fl07h8Lcriub76LL_O6vHFfhzFx2I2Oa-nD_W0-L7_qCan1ft1_XJXvd0uowWTHDjd_mML7W1u7Ha3vSje2umuR54juAMeZ2moGOOGplwqypQSOk2x0oJrIahUoeKKAaNGgwwUBEyFhgBhUtPEaAkttPrbPUmsSWy_lw2PDu38sZe5rC_zPKeYMjofgB9FR166
ClassificationCodes TP391.4
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 10.3778/j.issn.1002-8331.2404-0332
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
Discipline Engineering
DocumentTitle_FL YOLOv8 Crack Defect Detection Algorithm Based on Multi-Scale Features
EndPage 270
ExternalDocumentID jsjgcyyy202422023
GrantInformation_xml – fundername: (国家自然科学基金); (安徽省自然科学基金面上项目)
  funderid: (国家自然科学基金); (安徽省自然科学基金面上项目)
GroupedDBID -0Y
2B.
4A8
5XA
5XJ
92H
92I
93N
ABJNI
ACGFS
ALMA_UNASSIGNED_HOLDINGS
CCEZO
CUBFJ
CW9
PSX
TCJ
TGT
U1G
U5S
ID FETCH-LOGICAL-s1063-74b9d447f2b7cd24dd8ebb0de87e882cd9d7d4342fe3c5d354d9f1314ce2afec3
ISSN 1002-8331
IngestDate Thu May 29 04:10:55 EDT 2025
IsPeerReviewed false
IsScholarly false
Issue 22
Keywords 井壁裂缝
注意力机制
shaft lining crack
attention mechanism
目标检测
YOLOv8
object detection
轻量化
lightweight
Language Chinese
LinkModel OpenURL
MergedId FETCHMERGED-LOGICAL-s1063-74b9d447f2b7cd24dd8ebb0de87e882cd9d7d4342fe3c5d354d9f1314ce2afec3
PageCount 10
ParticipantIDs wanfang_journals_jsjgcyyy202422023
PublicationCentury 2000
PublicationDate 2024-11-15
PublicationDateYYYYMMDD 2024-11-15
PublicationDate_xml – month: 11
  year: 2024
  text: 2024-11-15
  day: 15
PublicationDecade 2020
PublicationTitle 计算机工程与应用
PublicationTitle_FL Computer Engineering and Applications
PublicationYear 2024
Publisher 安徽理工大学 电气与信息工程学院,安徽 淮南 232001%安徽理工大学 人工智能学院,安徽 淮南 232001
Publisher_xml – name: 安徽理工大学 电气与信息工程学院,安徽 淮南 232001%安徽理工大学 人工智能学院,安徽 淮南 232001
SSID ssib051375739
ssib001102935
ssj0000561668
ssib023646291
ssib057620132
Score 2.005659
Snippet TP391.4; 针对井壁裂缝背景复杂,纵横比差异大,导致检测效率低、漏检等问题,提出了一种融合多尺度特征的裂缝缺陷检测模型EDG-YOLO.设计特征提取模块EIRBlock(efficient...
SourceID wanfang
SourceType Aggregation Database
StartPage 261
Title 融合多尺度特征的YOLOv8裂缝缺陷检测算法
URI https://d.wanfangdata.com.cn/periodical/jsjgcyyy202422023
Volume 60
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVEBS
  databaseName: Inspec with Full Text
  issn: 1002-8331
  databaseCode: ADMLS
  dateStart: 20200501
  customDbUrl:
  isFulltext: true
  dateEnd: 99991231
  titleUrlDefault: https://www.ebsco.com/products/research-databases/inspec-full-text
  omitProxy: false
  ssIdentifier: ssib057620132
  providerName: EBSCOhost
link http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpR3LahRBsImbix7EJ74JYp_CxJl-THcfZ3YnBEnMwQTiKey8EnJYwSRCclLxoIjgRcSAXiIeRRARQ_BvNhv_wqqe3peJECMsTVHTU1OPnqrqYauLkFtlqFnOm76Xmrz0wOFxD7ZBMKSwWRGAVBl-Gpi5G07NizsLcmGkpgf-tbS-lk5km4fWlRzHqoADu2KV7D9YtkcUEACDfWEEC8N4JBvTRFOTUN2giaTGp1ojEAlqIgRin8YVENEopImi2tDYWAzcFSAGZmpxf3Z69pFGahGnmiE-rlPTcAASMdTAvYomoZ3jIxBLqmOcEyXU2Esxp1Uzy26-a2kmNAqGppl6lzFASntJW1KCxgAkjmcjLIcCrnbXBdKDx8IP5zZQRARAvkZ_So8c3Byggiq6sT9ERaASkBkQjSNGA5X64FcQJrAcsKoDtevWajdBLVYqjK2GUPHhgDTWApGyQAMUP-6kQKZD5Bjl6kkKwKTVT4gWiSYPU4sj5KwQMXgvDrIybqlbLQMK1oRR45DI2h4J8ticW-UOGctqDDHAG7eE_pO9gbhoA2e3uM4FzqoRhHMQVXV5NwxWB-y7jIpVrWH-DNZcKW2DNT5hoveECcgxhedz99l7-DD0ldWVpWxjYwOXAIOBnyCjDOK5XyOjUWNm-l5_KwCZs-lvBbBPQsj650LJgCup-ifiSiTiTjF1fQHCIHSVtI6z6sxiZPv235m2xX-tstlaGshT586Q026DORZV3uIsGdlcPkdODRw7ep6E-x9etV8_b3_can_Zae986rz40f75pLP1rPID-9tPO7vvO7s7v95939t-vPftZefz272vby6Q-clkrj7luf4p3moAOw9Poa8VQpUsVVnORJ7rIk39vNCqgI11loMnzgUXrCx4JnMuRW7KgAcCmwSWRcYvklrrQau4RMYKwdIskwWkvKFQkjUh8MtmmedNnxelCi-Tm07qRecfVxcP2OrKUSZdJSf77_c1Ult7uF5ch7x_Lb3hTPwbycG28w
linkProvider EBSCOhost
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=%E8%9E%8D%E5%90%88%E5%A4%9A%E5%B0%BA%E5%BA%A6%E7%89%B9%E5%BE%81%E7%9A%84YOLOv8%E8%A3%82%E7%BC%9D%E7%BC%BA%E9%99%B7%E6%A3%80%E6%B5%8B%E7%AE%97%E6%B3%95&rft.jtitle=%E8%AE%A1%E7%AE%97%E6%9C%BA%E5%B7%A5%E7%A8%8B%E4%B8%8E%E5%BA%94%E7%94%A8&rft.au=%E8%B5%B5%E4%BD%B0%E4%BA%AD&rft.au=%E7%A8%8B%E7%91%9E%E4%B8%B0&rft.au=%E8%B4%BE%E6%99%93%E8%8A%AC&rft.date=2024-11-15&rft.pub=%E5%AE%89%E5%BE%BD%E7%90%86%E5%B7%A5%E5%A4%A7%E5%AD%A6+%E7%94%B5%E6%B0%94%E4%B8%8E%E4%BF%A1%E6%81%AF%E5%B7%A5%E7%A8%8B%E5%AD%A6%E9%99%A2%2C%E5%AE%89%E5%BE%BD+%E6%B7%AE%E5%8D%97+232001%25%E5%AE%89%E5%BE%BD%E7%90%86%E5%B7%A5%E5%A4%A7%E5%AD%A6+%E4%BA%BA%E5%B7%A5%E6%99%BA%E8%83%BD%E5%AD%A6%E9%99%A2%2C%E5%AE%89%E5%BE%BD+%E6%B7%AE%E5%8D%97+232001&rft.issn=1002-8331&rft.volume=60&rft.issue=22&rft.spage=261&rft.epage=270&rft_id=info:doi/10.3778%2Fj.issn.1002-8331.2404-0332&rft.externalDocID=jsjgcyyy202422023
thumbnail_s http://utb.summon.serialssolutions.com/2.0.0/image/custom?url=http%3A%2F%2Fwww.wanfangdata.com.cn%2Fimages%2FPeriodicalImages%2Fjsjgcyyy%2Fjsjgcyyy.jpg