UAV-Based Structure from Motion – Multi-View Stereo (SfM-MVS): Mapping Cliff Face of Central Crater of Mt. Miharayama, Izu Oshima, Central Japan

Abstract

The purpose of this study is to develop a real-time measurement method for monitoring or detecting volcanic disasters using unmanned aerial vehicle (UAV). This study was supported by the Ministry of Education, Culture, Sports, Science and Technology (MEXT) Japan under “Integrated Program for Next Generation Volcano Research and Human Resource Development” project. Generally, the measurement of active volcano is dangerous because of less accessibility. But with the use of UAV human damage can be avoided compared to human exposure to danger, especially when manned aerial vehicle are used; furthermore, high-resolution imagery can be acquired quickly, which is better than satellite imagery. In general, aerial measurement is carried out by flying UAV horizontally and scanning the objects from nadir angle. However, measurement accuracy will be low since a camera cannot obtain enough data due to a big gradient caused by the crater wall. In this study, the UAV scanned inside crater from a direction approximately perpendicular to crater wall. The scanned images were used to create a 3D model of the interior of the crater. In addition, the correspondence between altitude of lava lake surface and crater volume was calculated. Therefore, it is possible to forecast lava overflow time based on the observed height of lava lake surface only.