3D POINT CLOUD BASED ON STRUCTURE-FROM-MOTION PHOTOGRAMMETRY FOR BRIDGE INCLINATION ASSESSMENT IN MYANMAR

摘要

Rapid urban expansion and economic development increases the demand for transport infrastructure in many developing countries. As of 2012, Myanmar has total of 4,728 bridge of which 741 bridge were classified as major bridge. Considering Myanmar vast geography, lack of technological resources and investment caused inadequate and inappropriate monitoring of bridge health. Structure-from-Motion (SfM) photogrammetry provides an economic and user-friendly alternative. Although SfM has been applied in various studies related to infrastructure, its application for bridge structure displacement remains limited. This study aimed to evaluate the application of SfM based reconstructed point cloud in assessing inclination of Twantay bridge's main towers, a suspension bridge located in suburban Yangon, Myanmar. Photosets were obtained around the towers at ground level using a consumer-grade digital camera. These images were processed with SfM workflow on commercial software and the result was then scaled based on known length of object features. A 3D reference point cloud was created using mesh model digitized from blueprint dataset to validate displacement from designed position. Boundary points at the edge of SfM reconstructed cloud were used for calculating distance difference to the corresponding nearest points on the reference cloud in x, y, and z direction. The result showed that the distance difference on the y-axis, a direction facing the river front, gradually increases toward the top part of the tower compared to that of bridge piers. This result suggested that SfM technology can be applied for inclination assessment to provide displacement tendency. However, precise measurement remains a challenge.