Improvement of GPR-Based Rebar Diameter Estimation Using YOLO-v3

As the frequency of earthquakes has increased in Korea recent years, designing earthquake-resistant facilities been increasingly emphasized. Structures constructed with rebars are vulnerable to shaking, which reduces their seismic performance and may result damage human life property. Because construction requires maintenance sub-constructions, such as by cutting or compensating for missing rebars, information on rebar diameter is required. In this study, YOLO-v3 algorithm, fastest object recognition performance, was applied structural correction data, a basic experiment conducted air predict facility, real time based ground-penetrating radar data. The reason using algorithm that case GPR data change slightly according reinforcing bar, it difficult discriminate naked eye, depending inspector. model achieved higher accuracy than conventional detection prediction methods. addition, possibility real-time during construction, proposed method, verified.