Three-Dimensional Reconstruction of Road Structural Defects Using GPR Investigation and Back-Projection Algorithm

Area of Science:

• Geophysics
• Civil Engineering
• Non-Destructive Testing

Background:

• Ground-Penetrating Radar (GPR) offers efficient, safe, and non-destructive road defect detection.
• Accurate characterization of road structural defects is crucial for infrastructure maintenance.

Purpose of the Study:

• To develop and validate a novel 3D reconstruction method for GPR images to precisely determine road structural defect parameters.
• To enhance the accuracy of non-destructive road inspection techniques.

Main Methods:

• Preprocessing of GPR images, including direct wave removal and wavelet denoising.
• Application of the back-projection (BP) algorithm for defect localization and sizing.
• 3D data set construction via interpolation and effective reflection data extraction using a clustering algorithm.
• 3D imaging performed in MATLAB, validated with gprMax simulations and laboratory models.

Main Results:

• The proposed method successfully reconstructed 3D models of road structural defects.
• Validation demonstrated the efficacy of the BP algorithm and clustering for accurate defect parameter determination.
• A correlation of approximately 0.67 was achieved between reconstructed and actual defects.

Conclusions:

• The developed 3D reconstruction method accurately identifies and characterizes road structural defects using GPR data.
• This technique significantly improves non-destructive road inspection capabilities.
• The study confirms the potential of integrated GPR processing and 3D reconstruction for infrastructure assessment.