Detecting Change for Multi-View, Long-Term Surface Inspection

Simon Stent, Riccardo Gherardi, Björn Stenger and Roberto Cipolla

Abstract

We describe a system for the detection of changes in multiple views of a tunnel surface. From data gathered by a robotic inspection rig, we use a structure-from-motion pipeline to build panoramas of the surface and register images from different time instances. Reliably detecting changes such as hairline cracks, water ingress and other surface damage between the registered images is a challenging problem: achieving the best possible performance for a given set of data requires sub-pixel precision and careful modelling of the noise sources. The task is further complicated by factors such as unavoidable registration error and changes in image sensors, capture settings and lighting. Our contribution is a novel approach to change detection using a two-channel convolutional neural network. The network accepts pairs of approximately registered image patches taken at different times and classifies them to detect anomalous changes. To train the network, we take advantage of synthetically generated training examples and the homogeneity of the tunnel surfaces to virtually eliminate the requirement for manual labelling. We evaluate our method on field data gathered from a live tunnel over several months, demonstrating it to outperform existing approaches from recent literature and industrial practice.

Session

Poster 2

Files

Extended Abstract (PDF, 2M)

Paper (PDF, 1227K)

DOI

10.5244/C.29.127
https://dx.doi.org/10.5244/C.29.127

Citation

Simon Stent, Riccardo Gherardi, Björn Stenger and Roberto Cipolla. Detecting Change for Multi-View, Long-Term Surface Inspection. In Xianghua Xie, Mark W. Jones, and Gary K. L. Tam, editors, Proceedings of the British Machine Vision Conference (BMVC), pages 127.1-127.12. BMVA Press, September 2015.

Bibtex

@inproceedings{BMVC2015_127,
	title={Detecting Change for Multi-View, Long-Term Surface Inspection},
	author={Simon Stent and Riccardo Gherardi and Björn Stenger and Roberto Cipolla},
	year={2015},
	month={September},
	pages={127.1-127.12},
	articleno={127},
	numpages={12},
	booktitle={Proceedings of the British Machine Vision Conference (BMVC)},
	publisher={BMVA Press},
	editor={Xianghua Xie, Mark W. Jones, and Gary K. L. Tam},
	doi={10.5244/C.29.127},
	isbn={1-901725-53-7},
	url={https://dx.doi.org/10.5244/C.29.127}
}