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.
Highlights
Civil structures such as bridges are reaching their end of service life due to aging, increased usage, and adverse climate impact [14]
We investigate the capabilities of transfer learning in the area of structural health monitoring
We find that the cross-domain transfer of pre-trained models proves helpful only when they are fine-tuned on the target task
Summary
Civil structures such as bridges are reaching their end of service life due to aging, increased usage, and adverse climate impact [14]. UAVs are cost-effective and safe for harder-toreach areas, they capture both damaged and structurally sound parts of bridges. Manual damage identification from such an enormous data source demands tremendous efforts and is prone to discrepancies due to human errors, fatigue, and poor judgments of bridge inspectors [2, 46]. The involved subjectivity in a visual inspection process results in inaccurate outcomes and poses serious concerns for public safety [40], as shown by the collapsing incidents like the Malahide viaduct [50] or the I-35 Minneapolis bridge [63]. We introduce related studies about automatic damage detection using deep (transfer) learning. We highlight studies about possible limitations of generic feature transfer using pre-trained models. Since pre-training on a large dataset is computationally expensive and time-consuming, several pretrained models have been made publicly available by academics and industry
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.