Compression-based unsupervised clustering of spectral signatures

This paper proposes to use compression-based similarity measures to cluster spectral signatures on the basis of their similarities. Such universal distances estimate the shared information between two objects by comparing their compression factors, which can be obtained by any standard compressor. Experiments on rocks categorization show that these methods may outperform traditional choices for spectral distances based on vector processing.

[1]  A. Gershman,et al.  Spectral matching accuracy in processing hyperspectral data , 2005, International Symposium on Signals, Circuits and Systems, 2005. ISSCS 2005..

[2]  R Muller,et al.  The processing chain and Cal/Val operations of the future hyperspectral satellite mission EnMAP , 2010, 2010 IEEE Aerospace Conference.

[3]  S. Hook,et al.  The ASTER spectral library version 2.0 , 2009 .

[4]  Paul M. B. Vitányi,et al.  Clustering by compression , 2003, IEEE Transactions on Information Theory.

[5]  S. M. de Jong,et al.  Imaging spectrometry : basic principles and prospective applications , 2001 .

[6]  F. Meer The effectiveness of spectral similarity measures for the analysis of hyperspectral imagery , 2006 .

[7]  Mihai Datcu,et al.  Algorithmic Information Theory-Based Analysis of Earth Observation Images: An Assessment , 2010, IEEE Geoscience and Remote Sensing Letters.

[8]  Chein-I. Chang,et al.  New Hyperspectral Discrimination Measure for Spectral Characterization , 2004 .

[9]  Alfonso Ortega,et al.  Common Pitfalls Using the Normalized Compression Distance: What to Watch Out for in a Compressor , 2005, Commun. Inf. Syst..

[10]  Eamonn J. Keogh,et al.  Towards parameter-free data mining , 2004, KDD.

[11]  Abraham Lempel,et al.  Compression of individual sequences via variable-rate coding , 1978, IEEE Trans. Inf. Theory.

[12]  Manuel Cebrián,et al.  The Normalized Compression Distance Is Resistant to Noise , 2007, IEEE Transactions on Information Theory.

[13]  Fred A. Kruse,et al.  The Spectral Image Processing System (SIPS) - Interactive visualization and analysis of imaging spectrometer data , 1993 .

[15]  P. R. Meneses,et al.  Spectral Correlation Mapper ( SCM ) : An Improvement on the Spectral Angle Mapper ( SAM ) , 2000 .