드론 촬영 정사영상을 이용한 합성곱 신경망 딥러닝 알고리즘 기반 콘크리트 균열 자동평가 시스템

다국어 초록 (Multilingual Abstract)

Cracks in concrete structure can lead to major structure failure if not noticed correctly at the right time. Various methods have been studied to detect concrete cracks due to the disadvantage of existing visual inspection. As the fourth industrial revolution has also affected the construction industry, concrete crack detection technologies using fields such as drones and AI are also developing. Vision recognition-based systems still have limitations such as identifying the location of cracks. This study proposes an efficient method of automatically detecting concrete cracks by using Convolutional Neural Network(CNN) with the Orthomosaic image, modeled with the help of UAV. The CNN model was developed and trained from combination of open source and original images.