Hyperspectral image clustering is an important and challenging problem, which aims to group image pixels according to the land cover information extracted from the spectrum. The spectrum observed at adjacent pixels are often highly-correlated, and leveraging such spatial correlation can greatly improve the clustering accuracy. Markov Random Field (MRF) is a powerful model to characterize such correlation. However, in this model the spatial parameter \beta often needs to be manually tuned, which brings difficulty in finding an optimal value. In this paper, we propose a novel hyperspectral clustering algorithm, which is able to learn parameter \beta from the data and thus achieves better performance. Specifically, we model the spectral information with Gaussian mixture model, and use variational expectation maximization method to complete the parameter estimation and clustering task. Experiments on both synthetic and real-world data sets verify the effectiveness of the proposed algorithm.

Hyperspectral Image clustering; undirected graphical model; variational expectation maximization,; spatial parameter