Land Cover Classification

Land cover classification aims to automatically categorize Earth's surface features from remotely sensed imagery, primarily to monitor environmental changes and manage resources. Current research heavily utilizes deep learning, focusing on convolutional neural networks (CNNs), transformers, and graph convolutional networks (GCNs) to improve classification accuracy, particularly for high-resolution imagery and multi-modal data (e.g., combining hyperspectral and LiDAR data). Significant efforts address challenges like data scarcity, noise in input data, and efficient handling of large datasets, with a focus on developing robust and computationally efficient models for large-scale applications. These advancements are crucial for improving environmental monitoring, precision agriculture, and urban planning.