Digital Surface Model

Digital surface models (DSMs) are 3D representations of the Earth's surface, crucial for various applications including solar energy potential mapping and urban planning. Current research focuses on improving DSM generation from satellite imagery, addressing challenges like low resolution, data sparsity, and noise through advanced techniques such as deep learning (including neural radiance fields and residual networks) and efficient registration algorithms. These improvements enhance the accuracy and scalability of DSM creation, leading to more precise analyses across diverse fields and facilitating applications ranging from renewable energy assessment to infrastructure development.