Real-time Assessment of Asphalt Pavement Internal Conditions Based on Automatic Signal Processing Using Ground-Penetrating Radar

Siqi Wang,Tao Ma,Xiaoming Huang
DOI: https://doi.org/10.5194/egusphere-egu24-7634
2024-01-01
Abstract:Ground-penetrating radar (GPR) has been widely used in asphalt pavement internal condition assessment, such as thickness measurement and defect detection. It requires identifying reflection features from the subsurface target, which could be accomplished via manual inspection. This may be subjective and lead to inaccurate detection results due to the field data noise. The wide electromagnetic footprint of air-coupled antenna, and strong pavement layer reflections could also camouflage reflection features of small-scale defects using GPR, e.g., concealed cracks. This article proposes an automatic method for asphalt pavement layer thickness determination and concealed crack detection using GPR. The effects of in-situ noise levels on detection accuracy were discussed. For thickness estimation, the interface reflections were located automatically based on a pipeline, including the Canny edge detector, wavelet transform, and matched filter. The vibration effect in the predicted thickness profile was removed using the empirical mode decomposition (EMD) method. The proposed method was validated using in-situ cores from the field test. For concealed crack detection, an edge detection-based method is proposed to enhance the reflections. A simulation model was built to investigate the antenna footprint and reflection features of concealed cracks in the asphalt pavement layer. Laboratory tests were performed to validate the proposed enhancement method. Results show that the proposed methods are automatic and can be implemented using existing programming functions, which may enable real-time asphalt pavement internal condition assessment during GPR surveys.
What problem does this paper attempt to address?