[CS10-20] A Study on damage detection of tunnel by semantic segmentation
Keywords:semantic segmentation, deep learning, damage detection, AI, CNN, ResNet
Out company is developing a tunnel inspection vehicle (MIMM-R) to streamline tunnel inspection. But further rationalization is required.
Based on the above, we studied the Semantic Segmentation that is one of the methods of the Deep Learning for the purpose of detecting damage (crack, peeling, free lime, water leakage) at the pixel level from the photographed image of MIMM-R.
As a result of verification, we evaluated that damage detection can be performed with appropriate accuracy although there is a problem for commercialization.
Based on the above, we studied the Semantic Segmentation that is one of the methods of the Deep Learning for the purpose of detecting damage (crack, peeling, free lime, water leakage) at the pixel level from the photographed image of MIMM-R.
As a result of verification, we evaluated that damage detection can be performed with appropriate accuracy although there is a problem for commercialization.
Abstract password authentication.
Password is required to view the abstract. Please enter a password to authenticate.