"Hyperspectral dataset"

"Hyperspectral remote sensing imagery plays a crucial role in various applications such as land cover classification, target detection, and environmental monitoring. To facilitate algorithm development and performance evaluation, several benchmark hyperspectral datasets have been widely adopted by the research community. Among the most commonly used datasets are Indian Pines, which captures agricultural scenarios with 220 spectral bands; Pavia University and Pavia Center, acquired over urban areas; Salinas, representing vegetation scenes; and Houston, which includes data from both airborne and spaceborne sensors. Additionally, datasets like Botswana, Kennedy Space Center (KSC), and WHU-Hi are frequently utilized for specific tasks. These datasets vary in spatial resolution, spectral range, and ground truth availability, providing diverse challenges for classification and unmixing algorithms. This abstract summarizes the key characteristics and typical applications of these widely used hyperspectral datasets."