|
| | Abstract | | We applied a supervised machine learning method for the purpose of auto detecting cavities under paved roadway from GPR Images. To detect cavity from massive data acquired by such like car borne GPR with high accuracy rapidly, automatic process based on knowledges of skilled-engineers is required, meanwhile automatic detection of cavities has not practically realized cause of its un-uniform response. We used actual GPR data which were acquired on natural occurred cavities as supervisors of machine learning method. At the verification with actual survey data, our method was able to detect cavity-caused GPR patterns, although there were still a lot of false detections. We think accumulation of labeled GPR data of cavity or the other will contribute improvement of our method. |
|
|
|
| |
|