Linear Feature Extraction for Hyperspectral Images Based on Information Theoretic Learning

被引：30

作者：

Kamandar, Mehdi ^{[1
]}

Ghassemian, Hassan ^{[1
]}

机构：

[1] Tarbiat Modares Univ, Fac Elect & Comp Engn, Tehran 141554843, Iran

来源：

IEEE GEOSCIENCE AND REMOTE SENSING LETTERS | 2013年 / 10卷 / 04期

关键词：

Hughes phenomenon; hyperspectral image classification; linear feature extractor; maximal relevance; minimal redundancy; CLASSIFICATION; REDUCTION;

D O I：

10.1109/LGRS.2012.2219575

中图分类号：

P3 [地球物理学]; P59 [地球化学];

学科分类号：

0708 ; 070902 ;

摘要：

This letter proposes a new supervised linear feature extractor for hyperspectral image classification. The criterion for feature extraction is a modified maximal relevance and minimal redundancy (MRMD), which has been used for feature selection until now. The MRMD is a function of mutual information terms, which possess higher order statistics of data; thus, it is effective for hyperspectral data with informative higher order statistics. The batch and stochastic versions of the gradient ascent are performed on the MRMD to find the optimal parameters of a linear feature extractor. Preliminary results achieve better classification performance than the traditional methods based on the first- and second-order moments of data.

引用

页码：702 / 706

页数：5

共 19 条

[11] Feature extraction using information-theoretic learning [J].

Hild, Kenneth E., II ;

Erdogmus, Deniz ;

Torkkola, Kari ;

Principe, Jose C. .

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2006, 28 (09) :1385-1392

[12] Hyperspectral data analysis and supervised feature reduction via projection pursuit [J].

Jimenez, LO ;

Landgrebe, DA .

IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 1999, 37 (06) :2653-2667

[13] Hyperspectral image data analysis [J].

Landgrebe, D .

IEEE SIGNAL PROCESSING MAGAZINE, 2002, 19 (01) :17-28

[14] A Composite Semisupervised SVM for Classification of Hyperspectral Images [J].

Marconcini, Mattia ;

Camps-Valls, Gustavo ;

Bruzzone, Lorenzo .

IEEE GEOSCIENCE AND REMOTE SENSING LETTERS, 2009, 6 (02) :234-238

[15] Classification of hyperspectral remote sensing images with support vector machines [J].

Melgani, F ;

Bruzzone, L .

IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2004, 42 (08) :1778-1790

[16] Feature selection based on mutual information: Criteria of max-dependency, max-relevance, and min-redundancy [J].

Peng, HC ;

Long, FH ;

Ding, C .

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2005, 27 (08) :1226-1238

[17] On locally adaptive density estimation [J].

Sain, SR ;

Scott, DW .

JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1996, 91 (436) :1525-1534

[18] THE EFFECT OF UNLABELED SAMPLES IN REDUCING THE SMALL SAMPLE-SIZE PROBLEM AND MITIGATING THE HUGHES PHENOMENON [J].

SHAHSHAHANI, BM ;

LANDGREBE, DA .

IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 1994, 32 (05) :1087-1095

[19] Independent component analysis-based dimensionality reduction with applications in hyperspectral image analysis [J].

Wang, Jing ;

Chang, Chein-I .

IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2006, 44 (06) :1586-1600

← 1 2 →