Concrete crack identification using computer vision in conjunction with machine learning
Concrete surface crack is one of the important indicator to check concrete durability and the associated problem.Although manual visual inspection is typically conducted to record crack-related information of length and width,this method is inefficient in the aspects of time,cost,and safety.Digital image processing is considered as an alternative of the manual visual inspection,however the discrimination of noncrack objects on concrete surface images is a critical issue for fully automatic identification.This paper presents a machine learning framework to automatically and accurately distinguish actual cracks from the image included both crack and noncrack.
Concrete crack concrete structure digital image processing machine learning
Hyunjun Kim Sung-Han Sim
School of Urban and Environmental Engineering,Ulsan National Institute of Science and Technology(UNIST),Ulsan,Republic of Korea
国际会议
The 7th World Conference on Structural Control and Monitoring(7WCSCM)(第七届结构控制与监测世界大会)
青岛
英文
1518-1519
2018-07-22（万方平台首次上网日期，不代表论文的发表时间）