Automatic Road Subsurface Distress Recognition from Ground Penetrating Radar Images using Deep Learning-based Cross-verification

Ground penetrating radar (GPR) has become a rapid and non-destructive solution for road subsurface distress (RSD) detection. Deep learning-based automatic RSD recognition, though ameliorating the burden of data processing, suffers from data scarcity and insufficient capability to recognize defects. In this study, a rigorously validated 3D GPR dataset containing 2134 samples of diverse types was constructed through field scanning. A novel cross-verification strategy was proposed to fully exploit the complementary abilities of region proposal networks in object recognition from different views of GPR images. The method achieves outstanding accuracy with a recall over 98.6% in field tests. The approach, integrated into an online RSD detection system, can reduce the human labor of inspection by around 90%.