Decoding the Moon's Surface: A Graph Neural Network Based Analysis of Chandrayaan-2 Lunar Data Classification

This study proposes a fused approach for geological classification of Chandrayaan-2 Imaging Infrared Spectrometer (IIRS) Dataset by combining deep learning and graph-based methods. Convolutional neural networks (CNNs) and Graph Convolutional Networks (GCNs) are utilized to extract features from the spectral information and capture spatial context, respectively. The results from both models are fused to enhance classification accuracy. The proposed approach addresses the challenges of high data dimensionality, limited training data, spectral range limitations, sensor noise, and complex lunar atmosphere. The effectiveness of the approach is evaluated using various accuracy measures.