{"title":"Dam surface crack detection based on deep learning","authors":"Linjing Li, Hua Zhang, Jie Pang, Jishuang Huang","doi":"10.1145/3366194.3366327","DOIUrl":null,"url":null,"abstract":"According to the statistics of the First National Water Census Bulletin in 2013[1], the number of water conservancy projects in China has exceeded 98,000, and 756 are under construction, with a total storage capacity of more than 930 billion cubic meter, ranking first in the world. While these water conservancy projects bring enormous economic and social benefits to China, they are affected by geology, hydrology, meteorology and other factors, and their buildings such as tunnels are prone to various defects. However, the current methods for detecting cracks on the dam surface are still dominated by humans. This process is not only inefficient, costly, but often incomplete. YOLOv2 lacks the capture of small defects, YOLOv3 uses three scale feature maps for prediction, and enhances the detection of small cracks. This paper aims to propose a new application scenario for applying YOLOv3 to crack detection in floodgate dam surface and share its effects.","PeriodicalId":105852,"journal":{"name":"Proceedings of the 2019 International Conference on Robotics, Intelligent Control and Artificial Intelligence","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2019-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"13","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of the 2019 International Conference on Robotics, Intelligent Control and Artificial Intelligence","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/3366194.3366327","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 13
Abstract
According to the statistics of the First National Water Census Bulletin in 2013[1], the number of water conservancy projects in China has exceeded 98,000, and 756 are under construction, with a total storage capacity of more than 930 billion cubic meter, ranking first in the world. While these water conservancy projects bring enormous economic and social benefits to China, they are affected by geology, hydrology, meteorology and other factors, and their buildings such as tunnels are prone to various defects. However, the current methods for detecting cracks on the dam surface are still dominated by humans. This process is not only inefficient, costly, but often incomplete. YOLOv2 lacks the capture of small defects, YOLOv3 uses three scale feature maps for prediction, and enhances the detection of small cracks. This paper aims to propose a new application scenario for applying YOLOv3 to crack detection in floodgate dam surface and share its effects.