A Weak-Signal-Aware Framework for Subsurface Defect Detection: Mechanisms for Enhancing Low-SCR Hyperbolic Signatures
This work addresses infrastructure inspection by improving detection of weak defects, though it appears incremental as it builds on existing lightweight detector methods.
The paper tackles the problem of detecting faint subsurface defects in Ground Penetrating Radar data, which are challenging due to low signal-to-clutter ratios, by proposing WSA-Net, a framework that enhances weak signals through physical-feature reconstruction, achieving 0.6958 mAP@0.5 and 164 FPS with 2.412 M parameters.
Subsurface defect detection via Ground Penetrating Radar is challenged by "weak signals" faint diffraction hyperbolas with low signal-to-clutter ratios, high wavefield similarity, and geometric degradation. Existing lightweight detectors prioritize efficiency over sensitivity, failing to preserve low-frequency structures or decouple heterogeneous clutter. We propose WSA-Net, a framework designed to enhance faint signatures through physical-feature reconstruction. Moving beyond simple parameter reduction, WSA-Net integrates four mechanisms: Signal preservation using partial convolutions; Clutter suppression via heterogeneous grouping attention; Geometric reconstruction to sharpen hyperbolic arcs; Context anchoring to resolve semantic ambiguities. Evaluations on the RTSTdataset show WSA-Net achieves 0.6958 mAP@0.5 and 164 FPS with only 2.412 M parameters. Results prove that signal-centric awareness in lightweight architectures effectively reduces false negatives in infrastructure inspection.