Use of the second-kind statistics for VHR SAR image retrieval

The Method of Log Cumulants (MoLC) has been recently proposed for analysis and parameter estimation of the probability density function (pdf) defined in R+. Here the Fourier transform as the characteristics function is replaced with Mellin transform for parameter estimation. Instead of focusing on pdf parameter estimation using MoLC, our objective in this article is to directly use log-moments (also called second-kind moments in second-kind statistics) of transformed synthetic aperture radar (SAR) images as the feature descriptor describing the image content. Gabor filters have been considered to transform the SAR image data. Comparison is presented with first and higher-order statistics and spectral feature descriptors. K-nearest neighborhood (K-NN) classification algorithm have been used as classifier.

[1]  Tao Li,et al.  Toward intelligent music information retrieval , 2006, IEEE Transactions on Multimedia.

[2]  F. Albregtsen.  Statistical Texture Measures Computed from Gray Level Coocurrence Matrices , 2008 .

[3]  Mihai Datcu,et al.  Contextual Descriptors for Scene Classes in Very High Resolution SAR Images , 2012, IEEE Geoscience and Remote Sensing Letters.

[4]  S. Quegan,et al.  Understanding Synthetic Aperture Radar Images , 1998 .

[5]  Mihai Datcu,et al.  Automated interpretation of very-high resolution SAR images , 2012, 2012 IEEE International Geoscience and Remote Sensing Symposium.

[6]  C. Oliver.  Optimum texture estimators for SAR clutter , 1993 .

[7]  Wen Hong,et al.  An Efficient and Flexible Statistical Model Based on Generalized Gamma Distribution for Amplitude SAR Images , 2010, IEEE Transactions on Geoscience and Remote Sensing.

[8]  Mihai Datcu,et al.  Selection of relevant features and TerraSAR-X products for classification of high resolution SAR images , 2012 .

[9]  Minh N. Do,et al.  Wavelet-based texture retrieval using generalized Gaussian density and Kullback-Leibler distance , 2002, IEEE Trans. Image Process..

[10]  Gabriele Moser,et al.  Dictionary-based stochastic expectation-maximization for SAR amplitude probability density function estimation , 2006, IEEE Transactions on Geoscience and Remote Sensing.

[11]  B. S. Manjunath,et al.  Texture Features for Browsing and Retrieval of Image Data , 1996, IEEE Trans. Pattern Anal. Mach. Intell..

[12]  John G. Proakis,et al.  Probability, random variables and stochastic processes , 1985, IEEE Trans. Acoust. Speech Signal Process..

[13]  Gabriele Moser,et al.  SAR amplitude probability density function estimation based on a generalized Gaussian model , 2006, IEEE Transactions on Image Processing.