GPR-Based Leakage Reconstruction of Shallow-Buried Water Supply Pipelines Using an Improved UNet++ Network
Open Access
- 25 June 2025
- journal article
- research article
- Published by MDPI AG in Remote Sensing
- Vol. 17 (13) , 2174
Abstract
Ground-penetrating radar (GPR) plays a critical role in detecting underground targets, particularly locating and characterizing leaks in buried pipelines. However, the complex nature of GPR images related to pipeline leaks, combined with the limitations of existing neural network-based inversion methods, such as insufficient feature extraction and low inversion accuracy, poses significant challenges for effective leakage reconstruction. To address these challenges, this paper proposes an enhanced UNet++-based model: the Multi-Scale Directional Network PlusPlus (MSDNet++). The network employs an encoder–decoder architecture, in which the encoder incorporates multi-scale directional convolutions with coordinate attention to extract and compress features across different scales effectively. The decoder fuses multi-level features through dense skip connections and further enhances the representation of critical information via coordinate attention, enabling the accurate inversion of dielectric constant images. Experimental results on both simulated and real-world data demonstrate that MSDNet++ can accurately invert the location and extent of buried pipeline leaks from GPR B-scan images.Keywords
Funding Information
- National Natural Science Foundation of China (62361015, 2025GXNSFAA069204)
- Guangxi Natural Science Foundation (62361015, 2025GXNSFAA069204)
This publication has 15 references indexed in Scilit:
- DMRF-UNet: A Two-Stage Deep Learning Scheme for GPR Data Inversion Under Heterogeneous Soil ConditionsIEEE Transactions on Antennas and Propagation, 2022
- Structural similarity index (SSIM) revisited: A data-driven approachExpert Systems with Applications, 2022
- GPRInvNet: Deep Learning-Based Ground-Penetrating Radar Data Inversion for Tunnel LiningsIEEE Transactions on Geoscience and Remote Sensing, 2021
- Deep Neural Network-Based Permittivity Inversions for Ground Penetrating Radar DataIEEE Sensors Journal, 2021
- UNet++: Redesigning Skip Connections to Exploit Multiscale Features in Image SegmentationIEEE Transactions on Medical Imaging, 2019
- Automatic Recognition of Highway Tunnel Defects Based on an Improved U-Net ModelIEEE Sensors Journal, 2019
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal NetworksIEEE Transactions on Pattern Analysis and Machine Intelligence, 2016
- Deep learningNature, 2015
- 2.5D forward and inverse modeling for interpreting low-frequency electromagnetic measurementsGeophysics, 2008
- A review of GPR for landmine detectionSensing and Imaging, 2006