Gaussian Mixture Models Clustering using Markov Random Field for Multispectral Remote Sensing Images

Multispectral images provide detailed data with information in both the spatial and spectral domains. Many clustering methods for multispectral images are based on a per-pixel classification, while uses only spectral information and ignores spatial information. In this work, a new clustering algorithm for multispectral images, based on both spectral and spatial information, is presented. This algorithm is integrated with generalized mixture Gaussian model (GMM) and Markov random field (MRF). The number of clusters is automatically identified by using the pseudolikelihood information criterion (PLIC). We examine the behavior of this model when applied to multispectral remote sensing images