Concrete crack detection and quantification using deep learning and structured light

被引:154
作者
Park, Song Ee [1 ]
Eem, Seung-Hyun [2 ]
Jeon, Haemin [1 ]
机构
[1] Hanbat Natl Univ, Dept Civil & Environm Engn, Daejeon 34158, South Korea
[2] Kyungpook Natl Univ, Sch Convergence & Fus Syst Engn, Sangju 37224, South Korea
基金
新加坡国家研究基金会;
关键词
Structural health monitoring; Crack; Detection; Quantification; Deep leaning; Structured light; IMPULSE-RESPONSE;
D O I
10.1016/j.conbuildmat.2020.119096
中图分类号
TU [建筑科学];
学科分类号
0813 ;
摘要
Considering the deterioration of civil infrastructures, the evaluation of structural safety by detecting cracks is becoming increasingly essential. In this paper, the advanced technologies of deep learning and structured light composed of vision and two laser sensors have been applied to detect and quantify cracks on surfaces of concrete structures. The YOLO (You Only Look Once) algorithm has been used for real-time detection, and the sizes of the detected cracks have been calculated based on the positions of the projected laser beams on the structural surface. Since laser beams may not be projected in parallel due to installation or manufacturing errors, the laser alignment correction algorithm with a specially designed jig module and a distance sensor is applied to increase the accuracy of the size measurement. The performance of the algorithm has been verified through simulations and experimental tests, and the results show that the cracks on the structural surfaces can be detected and quantified with high accuracy in real-time. (C) 2020 Elsevier Ltd. All rights reserved.
引用
收藏
页数:8
相关论文
共 27 条
[1]   Compressive strength evaluation of structural lightweight concrete by non-destructive ultrasonic pulse velocity method [J].
Alexandre Bogas, J. ;
Gloria Gomes, M. ;
Gomes, Augusto .
ULTRASONICS, 2013, 53 (05) :962-972
[2]  
[Anonymous], APPL INFRARED THERMO
[3]  
[Anonymous], APPL GEOGR GLOB FOR
[4]  
[Anonymous], INT C INT ROB APPL
[5]  
[Anonymous], HDB NONDESTRUCTIVE T
[6]   Assessing sensitivity of impact echo and ultrasonic surface waves methods for nondestructive evaluation of concrete structures [J].
Azari, Hoda ;
Nazarian, Soheil ;
Yuan, Deren .
CONSTRUCTION AND BUILDING MATERIALS, 2014, 71 :384-391
[7]  
Balageas D., 2010, Structural health monitoring
[8]   Deep learning-based automatic volumetric damage quantification using depth camera [J].
Beckman, Gustavo H. ;
Polyzois, Dimos ;
Cha, Young-Jin .
AUTOMATION IN CONSTRUCTION, 2019, 99 :114-124
[9]   Autonomous Structural Visual Inspection Using Region-Based Deep Learning for Detecting Multiple Damage Types [J].
Cha, Young-Jin ;
Choi, Wooram ;
Suh, Gahyun ;
Mahmoudkhani, Sadegh ;
Buyukozturk, Oral .
COMPUTER-AIDED CIVIL AND INFRASTRUCTURE ENGINEERING, 2018, 33 (09) :731-747
[10]   Deep Learning-Based Crack Damage Detection Using Convolutional Neural Networks [J].
Cha, Young-Jin ;
Choi, Wooram ;
Buyukozturk, Oral .
COMPUTER-AIDED CIVIL AND INFRASTRUCTURE ENGINEERING, 2017, 32 (05) :361-378