公路路基隐蔽性病害的探地雷达智能检测综述OA

As China's high-grade highway network progressively enters a peak maintenance period,the precise identification of hidden defects within the subgrade has become a core challenge for ensuring long-term pavement performance and traffic safety.Ground-penetrating radar(GPR),as an efficient non-destructive testing technology,enables non-destructive"imaging"of the internal subgrade structure,providing critical data for identifying typical defects such as cavities,voids,and water-rich areas.However,in complex real-world road environments,accurate interpretation of GPR data continues to face multiple difficulties.This paper systematically reviewed the research progress in intelligent GPR-based detection methods for highway subgrade hidden defects.It first analyzed the detection techniques for subgrade hidden defects and their current research status,elaborating on the basic principles of GPR and the application of various detection technologies.Subsequently,it summarized research methods for subgrade defect image processing and quantitative analysis.Finally,it focused on intelligent detection technologies based on deep learning,including the development and current application status of object detection and semantic segmentation in this field.This review aims to provide a theoretical reference and technical outlook for the intelligent identification and assessment of hidden defects in highway subgrade.