DEEP LEARNING-BASED INTERPRETATION METHOD OF BRIDGE POINT CLOUD DATA CONSIDERING UNCERTAINTY

Abstract

 In recent years, the utilization of 3D models in bridge management has been catching attention in order to improve the efficiency maintenance and the accuracy of damage diagnosis. There has been lots of research in which they construct 3D models using various data, including consturction drawings and point cloud data. However, there are some problems like the difficulty of applying their method to the vast number of existing bridges and the lack of accountability about the uncertainty and reliability of the models when they are constructed. In this research, the authors propose a 3D model construction method through the interpretation of bridge point cloud data using neural network learned generic shapes of bridge parts.