Parsimonious Gaussian process models for the classification of multivariate remote sensing images

A family of parsimonious Gaussian process models is presented. They allow to construct a Gaussian mixture model in a kernel feature space by assuming that the data of each class live in a specific subspace. The proposed models are used to build a kernel Markov random field (pGPMRF), which is applied to classify the pixels of a real multivariate remotely sensed image. In terms of classification accuracy, some of the proposed models perform equivalently to a SVM but they perform better than another kernel Gaussian mixture model previously defined in the literature. The pGPMRF provides the best classification accuracy thanks to the spatial regularization.

[1]  Lorenzo Bruzzone,et al.  Kernel methods for remote sensing data analysis , 2009 .

[2]  Jon Atli Benediktsson,et al.  Advances in Spectral-Spatial Classification of Hyperspectral Images , 2013, Proceedings of the IEEE.

[3]  Charles Bouveyron,et al.  Kernel discriminant analysis and clustering with parsimonious Gaussian process models , 2012, Statistics and Computing.

[4]  David A. Landgrebe,et al.  Toward an optimal supervised classifier for the analysis of hyperspectral data , 2004, IEEE Transactions on Geoscience and Remote Sensing.

[5]  Chein-I. Chang.  Hyperspectral Imaging: Techniques for Spectral Detection and Classification , 2003 .

[6]  Joachim M. Buhmann,et al.  On Relevant Dimensions in Kernel Feature Spaces , 2008, J. Mach. Learn. Res..

[7]  Stan Z. Li,et al.  Markov Random Field Modeling in Image Analysis , 2001, Computer Science Workbench.

[8]  Gunnar Rätsch,et al.  Input space versus feature space in kernel-based methods , 1999, IEEE Trans. Neural Networks.

[9]  John Platt,et al.  Probabilistic Outputs for Support vector Machines and Comparisons to Regularized Likelihood Methods , 1999 .

[10]  Jon Atli Benediktsson,et al.  SVM- and MRF-Based Method for Accurate Classification of Hyperspectral Images , 2010, IEEE Geoscience and Remote Sensing Letters.

[11]  Gabriele Moser,et al.  Combining Support Vector Machines and Markov Random Fields in an Integrated Framework for Contextual Image Classification , 2013, IEEE Transactions on Geoscience and Remote Sensing.