Hyperspectral data modelling by nonGaussian statistical distributions

In this manuscript we investigate on the statistical modeling of hyperspectral data. Accurately modeling real data is of paramount importance in the design of optimal classification or detection strategies and in evaluating their performances. In the work three nonGaussian models are considered and their capability in characterizing the statistical behavior of real data is discussed with reference to a data set acquired by the multispectral infrared and visible imaging spectrometer (MIVIS) sensor.

[1]  Marco Diani,et al.  An unsupervised algorithm for hyperspectral image segmentation based on the Gaussian mixture model , 2003, IGARSS 2003. 2003 IEEE International Geoscience and Remote Sensing Symposium. Proceedings (IEEE Cat. No.03CH37477).

[2]  A. Öztürk,et al.  Non-Gaussian random vector identification using spherically invariant random processes , 1993 .

[3]  Dimitris G. Manolakis,et al.  Detection algorithms for hyperspectral imaging applications , 2002, IEEE Signal Process. Mag..

[4]  John P. Kerekes,et al.  Statistics of hyperspectral imaging data , 2001, SPIE Defense + Commercial Sensing.

[5]  E. M. Winter,et al.  Anomaly detection from hyperspectral imagery , 2002, IEEE Signal Process. Mag..

[6]  T. Moon.  The expectation-maximization algorithm , 1996, IEEE Signal Process. Mag..

[7]  S. G. Beaven,et al.  Comparison of Gaussian mixture and linear mixture models for classification of hyperspectral data , 2000, IGARSS 2000. IEEE 2000 International Geoscience and Remote Sensing Symposium. Taking the Pulse of the Planet: The Role of Remote Sensing in Managing the Environment. Proceedings (Cat. No.00CH37120).

[8]  David A. Landgrebe,et al.  Signal Theory Methods in Multispectral Remote Sensing , 2003 .

[9]  Dimitris Manolakis,et al.  Non Gaussian models for hyperspectral algorithm design and assessment , 2002, IEEE International Geoscience and Remote Sensing Symposium.