Ensemble Space Carving for DSM Construction from Satellite Images

Digital Surface Models (DSMs) play a vital role in various fields, including urban planning, environmental monitoring, and 3D modeling, by providing height information of natural and man-made features on the Earth's surface. This paper presents a novel approach called Ensemble Space Carving for constructing DSMs directly from satellite imagery. By leveraging a combination of traditional computer vision techniques and machine learning, our method filters the images and amalgamates the resulting data. This innovative approach yields a simple yet robust DSM creation process without requiring feature point extraction, matching, and triangulation. The significance of DSMs in the specified field and the advantages of our proposed Ensemble Space Carving approach make this research a valuable contribution to satellite-based DSM generation.