Sparse representation-based augmented multinomial logistic extreme learning machine with weighted composite features for spectral-spatial classification of hyperspectral images

Faxian Cao, Zhijing Yang, Jinchang Ren, Wing-Kuen Ling, Huimin Zhao, Meijun Sun, Jon Atli Benediktsson

Research output: Contribution to journalArticle

8 Citations (Scopus)

Abstract

Although extreme learning machine (ELM) has successfully been applied to a number of pattern recognition problems, only with the original ELM it can hardly yield high accuracy for the classification of hyperspectral images (HSIs) due to two main drawbacks. The first is due to the randomly generated initial weights and bias, which cannot guarantee optimal output of ELM. The second is the lack of spatial information in the classifier as the conventional ELM only utilizes spectral information for classification of HSI. To tackle these two problems, a new framework for ELM-based spectral-spatial classification of HSI is proposed, where probabilistic modeling with sparse representation and weighted composite features (WCFs) is employed to derive the optimized output weights and extract spatial features. First, ELM is represented as a concave logarithmic-likelihood function under statistical modeling using the maximum a posteriori estimator. Second, sparse representation is applied to the Laplacian prior to efficiently determine a logarithmic posterior with a unique maximum in order to solve the ill-posed problem of ELM. The variable splitting and the augmented Lagrangian are subsequently used to further reduce the computation complexity of the proposed algorithm. Third, the spatial information is extracted using the WCFs to construct the spectral-spatial classification framework. In addition, the lower bound of the proposed method is derived by a rigorous mathematical proof. Experimental results on three publicly available HSI data sets demonstrate that the proposed methodology outperforms ELM and also a number of state-of-the-art approaches.
LanguageEnglish
Number of pages17
JournalIEEE Transactions on Geoscience and Remote Sensing
Early online date9 May 2018
DOIs
Publication statusE-pub ahead of print - 9 May 2018

Fingerprint

Learning systems
Logistics
logistics
Composite materials
machine learning
pattern recognition
modeling
Pattern recognition
Classifiers
methodology

Keywords

  • hyperspectral image (HSI)
  • spectral-spatial classification
  • extreme learning machine (ELM)
  • maximum a posterior (MAP)
  • sparse representation
  • Laplacian prior

Cite this

@article{1b92a6b9425149aa9df82dc288075f98,
title = "Sparse representation-based augmented multinomial logistic extreme learning machine with weighted composite features for spectral-spatial classification of hyperspectral images",
abstract = "Although extreme learning machine (ELM) has successfully been applied to a number of pattern recognition problems, only with the original ELM it can hardly yield high accuracy for the classification of hyperspectral images (HSIs) due to two main drawbacks. The first is due to the randomly generated initial weights and bias, which cannot guarantee optimal output of ELM. The second is the lack of spatial information in the classifier as the conventional ELM only utilizes spectral information for classification of HSI. To tackle these two problems, a new framework for ELM-based spectral-spatial classification of HSI is proposed, where probabilistic modeling with sparse representation and weighted composite features (WCFs) is employed to derive the optimized output weights and extract spatial features. First, ELM is represented as a concave logarithmic-likelihood function under statistical modeling using the maximum a posteriori estimator. Second, sparse representation is applied to the Laplacian prior to efficiently determine a logarithmic posterior with a unique maximum in order to solve the ill-posed problem of ELM. The variable splitting and the augmented Lagrangian are subsequently used to further reduce the computation complexity of the proposed algorithm. Third, the spatial information is extracted using the WCFs to construct the spectral-spatial classification framework. In addition, the lower bound of the proposed method is derived by a rigorous mathematical proof. Experimental results on three publicly available HSI data sets demonstrate that the proposed methodology outperforms ELM and also a number of state-of-the-art approaches.",
keywords = "hyperspectral image (HSI), spectral-spatial classification, extreme learning machine (ELM), maximum a posterior (MAP), sparse representation, Laplacian prior",
author = "Faxian Cao and Zhijing Yang and Jinchang Ren and Wing-Kuen Ling and Huimin Zhao and Meijun Sun and Benediktsson, {Jon Atli}",
note = "{\circledC} 2018 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting /republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.",
year = "2018",
month = "5",
day = "9",
doi = "10.1109/TGRS.2018.2828601",
language = "English",
journal = "IEEE Transactions on Geoscience and Remote Sensing",
issn = "0196-2892",
publisher = "IEEE",

}

Sparse representation-based augmented multinomial logistic extreme learning machine with weighted composite features for spectral-spatial classification of hyperspectral images. / Cao, Faxian; Yang, Zhijing; Ren, Jinchang; Ling, Wing-Kuen; Zhao, Huimin; Sun, Meijun; Benediktsson, Jon Atli.

In: IEEE Transactions on Geoscience and Remote Sensing, 09.05.2018.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Sparse representation-based augmented multinomial logistic extreme learning machine with weighted composite features for spectral-spatial classification of hyperspectral images

AU - Cao, Faxian

AU - Yang, Zhijing

AU - Ren, Jinchang

AU - Ling, Wing-Kuen

AU - Zhao, Huimin

AU - Sun, Meijun

AU - Benediktsson, Jon Atli

N1 - © 2018 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting /republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

PY - 2018/5/9

Y1 - 2018/5/9

N2 - Although extreme learning machine (ELM) has successfully been applied to a number of pattern recognition problems, only with the original ELM it can hardly yield high accuracy for the classification of hyperspectral images (HSIs) due to two main drawbacks. The first is due to the randomly generated initial weights and bias, which cannot guarantee optimal output of ELM. The second is the lack of spatial information in the classifier as the conventional ELM only utilizes spectral information for classification of HSI. To tackle these two problems, a new framework for ELM-based spectral-spatial classification of HSI is proposed, where probabilistic modeling with sparse representation and weighted composite features (WCFs) is employed to derive the optimized output weights and extract spatial features. First, ELM is represented as a concave logarithmic-likelihood function under statistical modeling using the maximum a posteriori estimator. Second, sparse representation is applied to the Laplacian prior to efficiently determine a logarithmic posterior with a unique maximum in order to solve the ill-posed problem of ELM. The variable splitting and the augmented Lagrangian are subsequently used to further reduce the computation complexity of the proposed algorithm. Third, the spatial information is extracted using the WCFs to construct the spectral-spatial classification framework. In addition, the lower bound of the proposed method is derived by a rigorous mathematical proof. Experimental results on three publicly available HSI data sets demonstrate that the proposed methodology outperforms ELM and also a number of state-of-the-art approaches.

AB - Although extreme learning machine (ELM) has successfully been applied to a number of pattern recognition problems, only with the original ELM it can hardly yield high accuracy for the classification of hyperspectral images (HSIs) due to two main drawbacks. The first is due to the randomly generated initial weights and bias, which cannot guarantee optimal output of ELM. The second is the lack of spatial information in the classifier as the conventional ELM only utilizes spectral information for classification of HSI. To tackle these two problems, a new framework for ELM-based spectral-spatial classification of HSI is proposed, where probabilistic modeling with sparse representation and weighted composite features (WCFs) is employed to derive the optimized output weights and extract spatial features. First, ELM is represented as a concave logarithmic-likelihood function under statistical modeling using the maximum a posteriori estimator. Second, sparse representation is applied to the Laplacian prior to efficiently determine a logarithmic posterior with a unique maximum in order to solve the ill-posed problem of ELM. The variable splitting and the augmented Lagrangian are subsequently used to further reduce the computation complexity of the proposed algorithm. Third, the spatial information is extracted using the WCFs to construct the spectral-spatial classification framework. In addition, the lower bound of the proposed method is derived by a rigorous mathematical proof. Experimental results on three publicly available HSI data sets demonstrate that the proposed methodology outperforms ELM and also a number of state-of-the-art approaches.

KW - hyperspectral image (HSI)

KW - spectral-spatial classification

KW - extreme learning machine (ELM)

KW - maximum a posterior (MAP)

KW - sparse representation

KW - Laplacian prior

U2 - 10.1109/TGRS.2018.2828601

DO - 10.1109/TGRS.2018.2828601

M3 - Article

JO - IEEE Transactions on Geoscience and Remote Sensing

T2 - IEEE Transactions on Geoscience and Remote Sensing

JF - IEEE Transactions on Geoscience and Remote Sensing

SN - 0196-2892

ER -