TriView-YOLO: Early Multi-View Fusion for Ground Penetrating Radar Cavity Detection in Soft, High-Water-Content Soils

2026-08-10

AI Summary

This study addresses the challenge of severe signal attenuation and poor void detectability in ground-penetrating radar (GPR) data acquired in high-moisture soft soils by proposing TriView-YOLO, a novel detection model based on the YOLOv12 architecture. The method introduces, for the first time, a nine-channel input comprising three orthogonal views—vertical B-scan, horizontal C-scan, and cross-sectional B-scan—and incorporates a custom TripleInputConv layer to enable early multi-view fusion, while producing detection bounding boxes exclusively from the vertical view. Trained on real-world data collected via a vehicle-mounted 3D GPR system, the model is evaluated under a rigorously designed protocol tailored to complex field conditions lacking public benchmarks. Experimental results demonstrate a mean average precision (mAP50) of 0.558 ± 0.028 on non-augmented test data, with per-image inference requiring only 3.1 ms (23.6 GFLOPs), significantly outperforming ablated variants and confirming the critical contribution of auxiliary views to detection performance.

Abstract

Automated detection of subsurface cavities from Ground Penetrating Radar (GPR) is most difficult in soft, high-water-content ground, where conductive, water-saturated soil attenuates the signal and degrades cavity reflections, yet this is also the condition under which cavities most readily form. This paper proposes TriView-YOLO, a multi-view YOLOv12 detector for road cavity screening in such ground. Three co-registered views (longitudinal B-scan, horizontal C-scan, and cross-section B-scan) form a 9-channel input fused by a TripleInputConv layer that replaces the YOLOv12 stem; the rest of the network is unchanged, and bounding boxes are required on the longitudinal view only. Training used 1,600 expert-verified field samples, principally metropolitan road surveys of Bangkok, Thailand, acquired with a vehicle-mounted multichannel three-dimensional GPR mobile mapping system, with surveys over the firmer subgrades of Japan added to training and validation only. The test set comes exclusively from the Bangkok surveys, over soft marine clay with 80-140% water content and a water table at 1-2 m depth, a ground condition for which no dedicated deep learning cavity-detection evaluation has been reported. On this unaugmented, field-only test set, split randomly within surveys, the proposed model attains mAP50 of 0.558 +/- 0.028 over three seeds at 23.6 GFLOPs and 3.1 ms per image. Ablations show that removing the auxiliary views lowers mAP50 and recall, whereas public and synthetic training images, DINOv3 features, larger model scale, and COCO pretraining bring no gain.

Problem

• Ground Penetrating Radar
• cavity detection
• high-water-content soils
• signal attenuation
• subsurface cavities

Innovation

• multi-view fusion
• Ground Penetrating Radar
• cavity detection
• YOLOv12
• TripleInputConv