Building displacement measurement and analysis based on UAV images: Fid-Bau Portal

Building displacement measurement and analysis based on UAV images

Sun, Jun / Peng, Bo / Wang, Cynthia Changxin / Chen, Kunxiang / Zhong, Botao / Wu, Jianjie

Abstract In engineering construction, displacement measurement of existing surrounding buildings is related to construction safety and quality. However, the conventional method using total station is slow and labour intensive, and more advanced laser scanning technology is expensive. To address these issues, a UAV-based approach is proposed to measure building displacement at selected marking points, which is contactless, rapid and low-cost. This approach consists of four steps: (1) the UAV image is transformed into point cloud; (2) the principle of region growth is applied to point cloud segmentation to obtain the marking points; (3) the coordinates of the marking points are obtained by combining the bounding box with a fitting plane; (4) the differences between the coordinates of the marking points at different times is the building displacement during this period. By comparing and analyzing the measurement results with a total station and 3D laser scanning technology, the approach's feasibility is verified.

Highlights • Introduces Unmanned Aerial Vehicles (UAV) image technology into the building displacement measurement. • Applies principle of regional growth to the point cloud segmentation to obtain the point cloud information of marking points. • Takes an optimization method to determine the center of marking points through the bounding box and a fitting plane. • Proposes a building displacement measurement method which is not only contactless but can also achieve rapid measurement. • Verifies the feasibility of the proposed method with the experiment of UAV, total station and 3D laser scanner.