Wei, L, Magee, DR and Cohn, AG orcid.org/0000-0002-7652-8907 (2018) An anomalous event detection and tracking method for a tunnel look-ahead ground prediction system. Automation in Construction, 91. pp. 216-225. ISSN 0926-5805
Abstract
The complicated geological conditions and unexpected geological hazards beyond the face of a tunnel are challenging problems for tunnel construction, which can cause great loss of life and property. While the geological surveys conducted before tunnel construction can provide rough information of construction site, they are not sufficiently accurate for predicting the sudden geological condition changes in local areas. Within the EU NETTUN project, an on-board ground prediction system consisting of multiple ground penetrating radars (GPR) and seismic sensors were developed to “see through” the ground and provide the local ground information behind the excavation front surface of a TBM (Tunnel Boring Machine). In order to facilitate the interpretation of the imaging data captured by this system, an automatic event detection and tracking method is presented in this paper. Anomalous 2D features are detected on each radar profile and reconstructed into a 3D accumulator; then, probable 3D events are detected from the accumulator and tracked at subsequent locations based on the information from multiple sets of radar data. The detection results can be used to generate alarms or be sent to human operators for interactive interpretation. The proposed method was evaluated using two sets of GPR data captured in a designed test field. Experimental results show that the buried targets can be correctly detected by the proposed event detection and tracking method. The proposed method is sufficiently flexible to cope with variations on the spatial configuration of on-board sensors.
Metadata
Item Type: | Article |
---|---|
Authors/Creators: |
|
Copyright, Publisher and Additional Information: | Copyright (c) 2018 Elsevier B. V. All rights reserved. This is an author produced version of a paper published in Automation in Construction. Uploaded in accordance with the publisher's self-archiving policy |
Keywords: | GPR data; Event detection; Tunnel construction; Ground prediction system |
Dates: |
|
Institution: | The University of Leeds |
Academic Units: | The University of Leeds > Faculty of Engineering & Physical Sciences (Leeds) > School of Computing (Leeds) |
Funding Information: | Funder Grant number EU - European Union 280712 EU - European Union 280712 EU - European Union 280712 EU - European Union 280712 |
Depositing User: | Symplectic Publications |
Date Deposited: | 13 Mar 2018 15:49 |
Last Modified: | 22 Mar 2019 01:43 |
Status: | Published |
Publisher: | Elsevier |
Identification Number: | 10.1016/j.autcon.2018.03.002 |
Open Archives Initiative ID (OAI ID): | oai:eprints.whiterose.ac.uk:128464 |