The process of construction site supervision plays a crucial role in ensuring safety and quality assurance in construction projects. However, traditional methods of supervision largely depend on human vision and individual experience, posing limitations in quickly detecting and preventing all defects. In particular, the thorough supervision of expansive sites is time-consuming and makes it challenging to identify all defects. This study proposes a new construction supervision system that utilizes vision processing technology and Artificial Intelligence(AI) to automatically detect and analyze defects as a solution to these issues. The system we developed is provided in the form of an application that operates on portable devices, designed to a lower technical barrier so that even non-experts can easily aid construction site supervision. The developed system swiftly and accurately identifies various potential defects at the construction site. As such, the introduction of this system is expected to significantly enhance the speed and accuracy of the construction supervision process.
본 연구는 한국과학재단이 주관하는 대학 중점연구소지원사업(No. NRF-2018R1A6A1A07025819)과 국토교통부의 디지털 기반 건축시공 및 안전감리 기술개발사업(1615012983)의 지원을 받아 수행되었습니다.