Unsupervised Machine Learning Framework for Identification of Spatial Distribution of Minerals on Mars

Planetary exploration missions have acquired a growing amount of remote sensing data, offering a reliable basis for studying the geological evolution of planetary bodies, such as Mars. In recent years, machine learning models have emerged as powerful tools for remote sensing by providing scalable and adaptive solutions for planetary science. In this study, we present an unsupervised machine learning framework for identifying and mapping the spatial distribution of minerals on Mars. Our framework utilises the Self-Organising Map (SOM) model and k-means clustering to identify clusters of spectral signatures, which may correspond to distinct minerals. Although the clusters can be labelled by referencing a spectral library, our framework does not require labelled data and can operate in an unsupervised manner. The framework retains full spectral dimensionality of input features while enabling topology-preserving clusters. The results indicate that our framework can identify the spatial distribution of minerals on Mars, even in complex spectral environments with overlapping features. Furthermore, the topological structure of mineral groupings that the SOM model preserves can provide important insights, including the indication of isolated minerals that may otherwise remain undetected in the target region.