The purpose of this study is to suggest a supervisory way to improve the efficiency of building Supervision using Deep Learning, especially object detecting technology. Since the establishment of the Building Supervision system, it has been changed and improved many times systematically, but it is hard to find any improvement in terms of implementation methods.
Therefore, the Building Supervision is until now the area where a lot of money, time and manpower are needed. It might give a room for superficial, formal and documentary supervision that could lead to faulty construction.
This study suggests a way of Building Supervision which is more automatic and effective so that it can lead to save the time, effort and money. And the way is to integrate BIM and Deep Learning-based object detection technology. On the assumption that BIM model is a modeling plan and a building should be built according to the plan, it was intended to identify whether the information of BIM and the information of the construction site were matched. And in this study, the subject of investigation were limited to the hoop-bar of the column. To develop this concept into technology, first we made a hoop-bar detecting Network by transfor learnning of YOLOv2 network through MATLAB. Among many training experiments, relatively most accurate network was selected, and second, we made BIM modeling through Revit according to the drawing and got the rebar quantity excel file from BIM model automatically. The last, we detected hoop-bar from a image with the network constructed from the first step and at the same time determine the consistency of detected hoop-bar quantity and BIM hoop-bar quantity through MATLAB. To prove its facility we applied this technology to building site in part and to improve detection accuracy we pre-processed the images. Through this technology we were able to detect hoop-bar for the 6 columns with more than 95% accuracy and also able to check for consistency within 8 seconds. The result showed the following possibilities.
First, the possibility of automated detection of rebar placement. Second, the possibility of integration of BIM and deep learning technology, and the last, the possibility of improvement of the efficiency of Building Supervision.