Damage detection using in-domain and cross-domain transfer learning

Zaharah Bukhsh (Corresponding author), Nils Jansen, Aaqib Saeed

Research output: Contribution to journalArticleAcademicpeer-review

30 Citations (Scopus)
79 Downloads (Pure)

Abstract

We investigate the capabilities of transfer learning in the area of structural health monitoring. In particular, we are interested in damage detection for concrete structures. Typical image datasets for such problems are relatively small, calling for the transfer of learned representation from a related large-scale dataset. Past efforts of damage detection using images have mainly considered cross-domain transfer learning approaches using pre-trained IMAGENET models that are subsequently fine-tuned for the target task. However, there are rising concerns about the generalizability of IMAGENET representations for specific target domains, such as for visual inspection and medical imaging. We, therefore, evaluate a combination of in-domain and cross-domain transfer learning strategies for damage detection in bridges. We perform comprehensive comparisons to study the impact of cross-domain and in-domain transfer, with various initialization strategies, using six publicly available visual inspection datasets. The pre-trained models are also evaluated for their ability to cope with the extremely low-data regime. We show that the combination of cross-domain and in-domain transfer persistently shows superior performance specially with tiny datasets. Likewise, we also provide visual explanations of predictive models to enable algorithmic transparency and provide insights to experts about the intrinsic decision logic of typically black-box deep models.
Original languageEnglish
Pages (from-to)16921-16936
Number of pages16
JournalNeural Computing and Applications
Volume33
Issue number24
Early online date13 Jul 2021
DOIs
Publication statusPublished - Dec 2021

Keywords

  • Cross-domain learning
  • Damage detection
  • In-domain learning
  • Pre-trained models
  • Transfer learning
  • Visual inspection

Fingerprint

Dive into the research topics of 'Damage detection using in-domain and cross-domain transfer learning'. Together they form a unique fingerprint.

Cite this