Detection of debondings with Ground Penetrating Radar using a machine learning method

摘要: In the field of civil engineering, Ground Penetrating Radar (GPR) is most widely used method Non-Destructive Testing (NDT). Using supervised learning methods or signal processing methods, it possible to analyze sub-surface defects in pavement. this paper, we propose use a machine called Support Vector Machines (SVM) detect presence debondings within Here, ground-coupled GPR quasi mono-static configuration along with SVM debondings. The experiments are done on bituminous concrete pavements various material characteristics. classification results show efficiency detection process.