Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
July 26, 2026Journal of the Korean society for railway

Physics-Based Deep Learning for Suppression of Reinforcing Steel Reflections in GPR: Application to Tunnel Lining Cavity Detection

View Full Paper
Ask AI
Bookmark
Share

Authors

SLSung Jin LeeRHRegidestyoko Wasistha HarsenoSKSeong‐Hoon Kee

Discussion

Loading...

Member takes

Overview

Randomized trial evaluates deep learning methods in ground penetrating radar for cavity detection, suggesting improved signal clarity.

Key Points

  • To enhance the visibility of cavity-related signals in ground penetrating radar (GPR) data impacted by rebar reflections.
  • Developed a physics-based deep learning framework using finite-difference time-domain modeling with gprMax.
  • Optimized parameters using the structural similarity index for better model accuracy in GPR data.
  • Generated synthetic B-scan images and trained U-Net, CycleGAN, and Pix2Pix models for performance evaluation.
  • Pix2Pix model achieved an SSIM of 0.97 and a PSNR of 31.31 dB, indicating superior noise suppression.
  • Validated model applied to actual GPR images from rebar-concrete tunnel lining, confirming improved visibility of cavity signals.
  • Framework enhances reliability and interpretability of GPR surveys under rebar interference conditions.

Cite This Study

Lee et al. (2026) studied this question.

synapsesocial.com/papers/6a65a306d3aea3239cd76421https://doi.org/10.7782/jksr.2026.29.7.708
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Simulation‐driven deep learning for rebar clutter elimination in ground‐penetrating radar images to detect backfill grout defects in segment linings2025
  2. 2Underground target detection method based on GPR physical mechanism constraints2026
  3. 3Study on Improving the Detection Accuracy of Non-metallic Underground Facilities using Frequency-Domain Data Fusion of HS GPR and RTS GPR2026
  4. 4Study on Vehicle-Mounted GPR Clutter Processing Algorithm2026
  5. 5GPR Noise Reduction Network Based on Multi-Domain Constrained TransUNet2026