COMPARATIVE ANALYSIS OF DIGITAL ELEVATION MODELS GENERATED WITH SFM (STRUCTURE-FROM-MOTION) USING AERIAL IMAGERY AND WITH LIDAR (LIGHT DETECTION AND RANGING) TECHNOLOGY

摘要

The Nationwide Disaster Risk and Exposure Assessment for Mitigation (DREAM) Program produces updated and high-resolution Digital Elevation Models (DEMs), maps, and orthophotographs of the Philippines' major river basins. These produced DEMs are used by the program in creating hazard models for disaster risk reduction and can also be used for other applications in varying fields such as engineering, urban and regional planning and resource management. In order to generate the DEMs, the program uses state-of-the-art Light Detection and Ranging (LiDAR) technology to produce maps of the covered areas in one meter resolution. This remote sensing method uses light in the form of pulsed lasers to measure distances to the surface of the Earth from an aerial platform which, when combined with other positioning information, generates precise, three-dimensional digital representations of the ground surface. In contrast, Structure from Motion (SfM) is a method developed by the photogrammetric and computer vision communities for digitally recreating objects and environments in three-dimensions from a series of overlapping but offset imagery. Comparative analysis of the DEMs produced by the two methods, one produced using classified LiDAR point cloud data and the other produced using SfM processing of DREAM's aerial images, has shown that SfM is capable of recreating the topography and features on the surface of the Earth. It was also observed that the accuracy of the generated DEMs through SfM was greatly affected by the topographic configuration of the ground being modelled.