Deep active learning for civil infrastructure defect detection and classification

Chen Feng, Ming Yu Liu, Chieh Chi Kao, Teng Yok Lee

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Automatic detection and classification of defects in infrastructure surface images can largely boost its maintenance efficiency. Given enough labeled images, various supervised learning methods have been investigated for this task, including decision trees and support vector machines in previous studies, and deep neural networks more recently. However, in real world applications, labels are harder to obtain than images, due to the limited labeling resources (i.e., experts). Thus we propose a deep active learning system to maximize the performance. A deep residual network is firstly designed for defect detection and classification in an image. Following our active learning strategy, this network is trained as soon as an initial batch of labeled images becomes available. It is then used to select a most informative subset of new images and query labels from experts to retrain the network. Experiments demonstrate more efficient performance improvements of our method than baselines, achieving 87.5% detection accuracy.

Original languageEnglish (US)
Title of host publicationComputing in Civil Engineering 2017
Subtitle of host publicationSmart Safety, Sustainability and Resilience - Selected Papers from the ASCE International Workshop on Computing in Civil Engineering 2017
EditorsKen-Yu Lin, Ken-Yu Lin, Nora El-Gohary, Nora El-Gohary, Pingbo Tang, Pingbo Tang
PublisherAmerican Society of Civil Engineers (ASCE)
Pages298-306
Number of pages9
ISBN (Electronic)9780784480823, 9780784480847
DOIs
StatePublished - 2017
Event2017 ASCE International Workshop on Computing in Civil Engineering, IWCCE 2017 - Seattle, United States
Duration: Jun 25 2017Jun 27 2017

Publication series

NameCongress on Computing in Civil Engineering, Proceedings
Volume0

Other

Other2017 ASCE International Workshop on Computing in Civil Engineering, IWCCE 2017
CountryUnited States
CitySeattle
Period6/25/176/27/17

ASJC Scopus subject areas

  • Civil and Structural Engineering
  • Computer Science Applications

Fingerprint Dive into the research topics of 'Deep active learning for civil infrastructure defect detection and classification'. Together they form a unique fingerprint.

  • Cite this

    Feng, C., Liu, M. Y., Kao, C. C., & Lee, T. Y. (2017). Deep active learning for civil infrastructure defect detection and classification. In K-Y. Lin, K-Y. Lin, N. El-Gohary, N. El-Gohary, P. Tang, & P. Tang (Eds.), Computing in Civil Engineering 2017: Smart Safety, Sustainability and Resilience - Selected Papers from the ASCE International Workshop on Computing in Civil Engineering 2017 (pp. 298-306). (Congress on Computing in Civil Engineering, Proceedings; Vol. 0). American Society of Civil Engineers (ASCE). https://doi.org/10.1061/9780784480823.036