Extreme sparse multinomial logistic regression: a fast and robust framework for hyperspectral image classification

Faxian Cao, Zhijing Yang, Jinchang Ren, Wing-Kuen Ling, Huimin Zhao, Stephen Marshall

Research output: Contribution to journalArticle

Abstract

Although the sparse multinomial logistic regression (SMLR) has provided a useful tool for sparse classification, it suffers from inefficacy in dealing with high dimensional features and manually set initial regressor values. This has significantly constrained its applications for hyperspectral image (HSI) classification. In order to tackle these two drawbacks, an extreme sparse multinomial logistic regression (ESMLR) is proposed for effective classification of HSI. First, the HSI dataset is projected to a new feature space with randomly generated weight and bias. Second, an optimization model is established by the Lagrange multiplier method and the dual principle to automatically determine a good initial regressor for SMLR via minimizing the training error and the regressor value. Furthermore, the extended multi-attribute profiles (EMAPs) are utilized for extracting both the spectral and spatial features. A combinational linear multiple features learning (MFL) method is proposed to further enhance the features extracted by ESMLR and EMAPs. Finally, the logistic regression via the variable splitting and the augmented Lagrangian (LORSAL) is adopted in the proposed framework for reducing the computational time. Experiments are conducted on two well-known HSI datasets, namely the Indian Pines dataset and the Pavia University dataset, which have shown the fast and robust performance of the proposed ESMLR framework.
LanguageEnglish
Number of pages22
JournalRemote Sensing
Volume9
Issue number12
DOIs
StatePublished - 2 Dec 2017

Fingerprint

Image classification
image classification
Logistics
logistics
Lagrange multipliers
learning
experiment
Experiments

Keywords

  • hyperspectral image (HSI) classification
  • sparse multinomial logistic regression (SMLR)
  • extreme sparse multinomial logistic regression (ESMLR)
  • extended multi-attribute profiles (EMAPs)
  • linear multiple features learning (MFL)
  • Lagrange multiplier

Cite this

@article{95b8a6bcfde947d5ae95a1a0ec7241f2,
title = "Extreme sparse multinomial logistic regression: a fast and robust framework for hyperspectral image classification",
abstract = "Although the sparse multinomial logistic regression (SMLR) has provided a useful tool for sparse classification, it suffers from inefficacy in dealing with high dimensional features and manually set initial regressor values. This has significantly constrained its applications for hyperspectral image (HSI) classification. In order to tackle these two drawbacks, an extreme sparse multinomial logistic regression (ESMLR) is proposed for effective classification of HSI. First, the HSI dataset is projected to a new feature space with randomly generated weight and bias. Second, an optimization model is established by the Lagrange multiplier method and the dual principle to automatically determine a good initial regressor for SMLR via minimizing the training error and the regressor value. Furthermore, the extended multi-attribute profiles (EMAPs) are utilized for extracting both the spectral and spatial features. A combinational linear multiple features learning (MFL) method is proposed to further enhance the features extracted by ESMLR and EMAPs. Finally, the logistic regression via the variable splitting and the augmented Lagrangian (LORSAL) is adopted in the proposed framework for reducing the computational time. Experiments are conducted on two well-known HSI datasets, namely the Indian Pines dataset and the Pavia University dataset, which have shown the fast and robust performance of the proposed ESMLR framework.",
keywords = "hyperspectral image (HSI) classification, sparse multinomial logistic regression (SMLR), extreme sparse multinomial logistic regression (ESMLR), extended multi-attribute profiles (EMAPs), linear multiple features learning (MFL), Lagrange multiplier",
author = "Faxian Cao and Zhijing Yang and Jinchang Ren and Wing-Kuen Ling and Huimin Zhao and Stephen Marshall",
year = "2017",
month = "12",
day = "2",
doi = "10.3390/rs9121255",
language = "English",
volume = "9",
journal = "Remote Sensing",
issn = "2072-4292",
number = "12",

}

Extreme sparse multinomial logistic regression : a fast and robust framework for hyperspectral image classification. / Cao, Faxian; Yang, Zhijing; Ren, Jinchang; Ling, Wing-Kuen; Zhao, Huimin; Marshall, Stephen.

In: Remote Sensing, Vol. 9, No. 12, 02.12.2017.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Extreme sparse multinomial logistic regression

T2 - Remote Sensing

AU - Cao,Faxian

AU - Yang,Zhijing

AU - Ren,Jinchang

AU - Ling,Wing-Kuen

AU - Zhao,Huimin

AU - Marshall,Stephen

PY - 2017/12/2

Y1 - 2017/12/2

N2 - Although the sparse multinomial logistic regression (SMLR) has provided a useful tool for sparse classification, it suffers from inefficacy in dealing with high dimensional features and manually set initial regressor values. This has significantly constrained its applications for hyperspectral image (HSI) classification. In order to tackle these two drawbacks, an extreme sparse multinomial logistic regression (ESMLR) is proposed for effective classification of HSI. First, the HSI dataset is projected to a new feature space with randomly generated weight and bias. Second, an optimization model is established by the Lagrange multiplier method and the dual principle to automatically determine a good initial regressor for SMLR via minimizing the training error and the regressor value. Furthermore, the extended multi-attribute profiles (EMAPs) are utilized for extracting both the spectral and spatial features. A combinational linear multiple features learning (MFL) method is proposed to further enhance the features extracted by ESMLR and EMAPs. Finally, the logistic regression via the variable splitting and the augmented Lagrangian (LORSAL) is adopted in the proposed framework for reducing the computational time. Experiments are conducted on two well-known HSI datasets, namely the Indian Pines dataset and the Pavia University dataset, which have shown the fast and robust performance of the proposed ESMLR framework.

AB - Although the sparse multinomial logistic regression (SMLR) has provided a useful tool for sparse classification, it suffers from inefficacy in dealing with high dimensional features and manually set initial regressor values. This has significantly constrained its applications for hyperspectral image (HSI) classification. In order to tackle these two drawbacks, an extreme sparse multinomial logistic regression (ESMLR) is proposed for effective classification of HSI. First, the HSI dataset is projected to a new feature space with randomly generated weight and bias. Second, an optimization model is established by the Lagrange multiplier method and the dual principle to automatically determine a good initial regressor for SMLR via minimizing the training error and the regressor value. Furthermore, the extended multi-attribute profiles (EMAPs) are utilized for extracting both the spectral and spatial features. A combinational linear multiple features learning (MFL) method is proposed to further enhance the features extracted by ESMLR and EMAPs. Finally, the logistic regression via the variable splitting and the augmented Lagrangian (LORSAL) is adopted in the proposed framework for reducing the computational time. Experiments are conducted on two well-known HSI datasets, namely the Indian Pines dataset and the Pavia University dataset, which have shown the fast and robust performance of the proposed ESMLR framework.

KW - hyperspectral image (HSI) classification

KW - sparse multinomial logistic regression (SMLR)

KW - extreme sparse multinomial logistic regression (ESMLR)

KW - extended multi-attribute profiles (EMAPs)

KW - linear multiple features learning (MFL)

KW - Lagrange multiplier

UR - http://www.mdpi.com/journal/remotesensing

U2 - 10.3390/rs9121255

DO - 10.3390/rs9121255

M3 - Article

VL - 9

JO - Remote Sensing

JF - Remote Sensing

SN - 2072-4292

IS - 12

ER -