Damage volumetric assessment and digital twin synchronization based on LiDAR point clouds

Point clouds are widely used for structure inspection and can provide damage spatial information. However, how to update a digital twin (DT) with local damage based on point clouds has not been sufficiently studied. This research presents an efficient framework for assessing and DT synchronizing local damage on a planar surface using point clouds. The pipeline starts from damage detection via DeepLabV3+ on the pseudo grayscale images from the point depth. It avoids the drawbacks of image and point cloud fusion. The target point cloud is separated according to the detected damage. Then, it can be converted into a 3D binary matrix through voxelization and binarization, which is highly lightweight and can be losslessly compressed for DT synchronization. The framework is validated via two case studies, demonstrating that the proposed voxel-based method can be easily applied to real-world damage with non-convex geometry instead of convex-hull fitting; finite-element (FE) models and BIM models can be updated automatically through the framework.