In recent years, typhoons and heavy rains in Taiwan are often accompanied by sediment disasters such as landslides and debris flows. If disaster area information can be obtained in the shortest possible period of time for rapid assessment of sediment transport and disaster scale, it can provide quantitative reference information for follow-up disaster relief and decision-making. Focusing on the Leye collapse in the Typhoon Morakot disaster area and its surroundings as the research area, this study employed UAV, which defies ground photography angle limits and performs low-altitude photography of the disaster-hit area. Next, by using unconventional photogrammetry technology, a field of computer vision technology, combined with VBS-RTK technology, which has become mature in recent years, this study aims to accomplish real time positioning of the reference control points and rapidly reconstruct the environmental 3D elevation data of the disaster area. The proposed approach can enable a rapid quantitative assessment of major sources of sediment, potential unstable areas, and the scale of the sediment disaster. Topographic data obtained using the proposed method can be applied to rapidly calculate landslide depth, and landslide volume after comparison with pre-disaster DTM data.