Defect segmentation: Mapping tunnel lining internal defects with ground penetrating radar data using a convolutional neural network. (14th February 2022)

• Record Type: Journal Article
• Authors: Yang, Senlin; Wang, Zhengfang; Wang, Jing; Cohn, Anthony G.; Zhang, Jiaqi; Jiang, Peng; Nie, Lichao; Sui, Qingmei
• Is Part Of: Construction & building materials. Volume 319(2022)
• Publication Date: 2022-02-14
• Subjects: Convolutional neural networks (CNNs) -- Ground Penetrating Radar (GPR) -- GPR data intelligent recognition -- Tunnel lining defect; Building materials -- Periodicals; 624.18
• DOI: 10.1016/j.conbuildmat.2021.125658
• Languages: English
• ISSNs: 0950-0618

Abstract

Abstract: This work offers a defect segmentation approach for the nondestructive testing of tunnel lining internal defects using Ground Penetrating Radar (GPR) data. Given GPR synthetic data, it maps the internal defect structure, using a CNN named Segnet coupled with the Lovász softmax loss function, which enhances the accuracy, automation, and efficiency of defect identification. Experiments with both synthetic and actual data show that our innovative method overcomes problems in standard GPR data interpretation. A physical test model with a known defect was developed and manufactured, and GPR data was acquired and analyzed to verify the approach. Highlights: Defect segmentation is our proposed tunnel lining defect detection. Inputting GPR directly into CNNs profiles internal lining defects. CNN correctly identified defect types and location, and achieved reliable results. Segnet was introduced to the defect segmentation method for more accurate results. Model building and data processing verified the proposed method.