Title
Anomaly detection of defects on concrete structures with the convolutional autoencoder.
Abstract
•Deep learning model is applied for the anomaly detection of concrete defects.•The model training is in the unsupervised mode, with no label needed.•This anomaly detection technique is adaptable to defects on wide ranges of scales.•The technique outperforms classical automatic methods in concrete defect detection.•Anomaly scores of the anomaly map alert inspectors for any potential defects.
Year
DOI
Venue
2020
10.1016/j.aei.2020.101105
Advanced Engineering Informatics
Keywords
DocType
Volume
Anomaly detection,Unsupervised learning,Convolutional autoencoder,Concrete structure,Cracking,Spalling
Journal
45
ISSN
Citations 
PageRank 
1474-0346
5
0.60
References 
Authors
0
6
Name
Order
Citations
PageRank
Jun Kang Chow150.94
Z. Su250.60
Junjie Wu355147.60
Pin Siang Tan450.60
X. Mao550.60
Y. H. Wang650.60