Classification of hyperspectral images by integrating spectral and spatial information

Hyperspectral imaging has emerged as a prominent technology for remotely sensing and analyzing complex landscapes. However, accurately classifying hyperspectral images is a challenging task due to the high-dimensionality and inherent spectral variability. In this study, we propose a novel approach to enhance classification accuracy by integrating both spectral and spatial information. The main objective of this research is to address the limitations of traditional approaches that rely solely on spectral information for classification. To achieve this, we applied a two-step approach. First, we extracted the spectral features from the hyperspectral images using state-of-the-art algorithms. Then, we integrated spatial information by considering the contextual relationships between pixels through spatial filtering techniques. This allowed us to capture more relevant information and enhance the classification performance. This study highlights the importance of incorporating both spectral and spatial information for the accurate classification of hyperspectral images.