Multitask GLocal OBIA-Mamba for Sentinel-2 Landcover Mapping

The Multitask Glocal OBIA-Mamba (MSOM) model for Sentinel-2 land cover mapping addresses significant challenges in environmental monitoring, as highlighted in related research on biophysical parameter estimation using Sentinel-2 imagery. The physics-informed Transformer-VAE model emphasizes the importance of accurate vegetation data for ecosystem management, paralleling the MSOM's focus on enhancing classification accuracy. Furthermore, the exploration of multimodal LLMs in visual information processing underscores the need for effective integration of diverse data types, a theme that resonates with the MSOM's dual-branch architecture. Together, these studies contribute to advancing methodologies in remote sensing and ecological monitoring.