A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Deep learning-based pavement subsurface distress detection via ground penetrating radar data
Abstract Pavement subsurface distress endangers driving safety and road serviceability. Ground penetrating radar (GPR) can non-destructively provides high-resolution profiles of road. However, the automatic interpretation of radar signals remains challenging. This study proposed an automatic pavement subsurface distress detection method using traditional signal processing and deep learning. Firstly, a piecewise linear function for radar signal automatic gain was proposed. Wavelet transform was applied to identify road layers for function segmentation. Each signal's power spectral density was calculated to determine the gain function coefficient. A specific pseudo-color mapping method was designed to convert reflected signal for deep learning model training. A radar reflection simulation was built to pre-train the model to enhance model performance. More than 270 km of field test data were collected for model training and validation. The proposed model has shown a 72.39% accuracy in shallow cracks and 68.74% in subsurface voids.
Highlights Pre-training the model on GPR simulated images can effectively improve the performance in few-shot training. The automatic gain method can effectively eliminate the attenuation of GPR reflected signal. Pseudo-color mapping can more fully retain the phase information, which is convenient for model learning and prediction.
Deep learning-based pavement subsurface distress detection via ground penetrating radar data
Abstract Pavement subsurface distress endangers driving safety and road serviceability. Ground penetrating radar (GPR) can non-destructively provides high-resolution profiles of road. However, the automatic interpretation of radar signals remains challenging. This study proposed an automatic pavement subsurface distress detection method using traditional signal processing and deep learning. Firstly, a piecewise linear function for radar signal automatic gain was proposed. Wavelet transform was applied to identify road layers for function segmentation. Each signal's power spectral density was calculated to determine the gain function coefficient. A specific pseudo-color mapping method was designed to convert reflected signal for deep learning model training. A radar reflection simulation was built to pre-train the model to enhance model performance. More than 270 km of field test data were collected for model training and validation. The proposed model has shown a 72.39% accuracy in shallow cracks and 68.74% in subsurface voids.
Highlights Pre-training the model on GPR simulated images can effectively improve the performance in few-shot training. The automatic gain method can effectively eliminate the attenuation of GPR reflected signal. Pseudo-color mapping can more fully retain the phase information, which is convenient for model learning and prediction.
Deep learning-based pavement subsurface distress detection via ground penetrating radar data
Li, Yishun (author) / Liu, Chenglong (author) / Yue, Guanghua (author) / Gao, Qian (author) / Du, Yuchuan (author)
2022-07-28
Article (Journal)
Electronic Resource
English
Interpreting Ground Penetrating Radar Pavement Data
| British Library Conference Proceedings | 1997
Deep learning pavement segregation identification method based on ground penetrating radar
| European Patent Office | 2023
Deep learning-based detection of tie bars in concrete pavement using ground penetrating radar
| Taylor & Francis Verlag | 2023