YOLO Based Bridge Surface Defect Detection Using Decoupled Prediction

Songyang Sun, Wei Liu, Rongxin Cui

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

6 Scopus citations

Abstract

Bridge surface defect detection is an essential method for evaluating the bridge quality and subsequent repair. Convolutional neural network based intelligent object detectors have been powerful for defect detection in recent years. In this paper, we propose a model based on the YOLO detector for detection. We use the convolutional block attention module, decoupled prediction head, and focal loss function to improve performance. To verify our proposed model, we perform experiments on an open bridge surface defect dataset, and our model can obtain 90.3% mAP50 and 72.8% mAP75. The detection precision of our network is higher than the original YOLOv5.

Original languageEnglish
Title of host publication2022 7th Asia-Pacific Conference on Intelligent Robot Systems, ACIRS 2022
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages117-122
Number of pages6
ISBN (Electronic)9781665485197
DOIs
StatePublished - 2022
Event7th Asia-Pacific Conference on Intelligent Robot Systems, ACIRS 2022 - Tianjin, China
Duration: 1 Jul 20223 Jul 2022

Publication series

Name2022 7th Asia-Pacific Conference on Intelligent Robot Systems, ACIRS 2022

Conference

Conference7th Asia-Pacific Conference on Intelligent Robot Systems, ACIRS 2022
Country/TerritoryChina
CityTianjin
Period1/07/223/07/22

Keywords

  • attention mechanism
  • decoupled prediction
  • defect detection
  • YOLO detector

Fingerprint

Dive into the research topics of 'YOLO Based Bridge Surface Defect Detection Using Decoupled Prediction'. Together they form a unique fingerprint.

Cite this