Classification of hyperspectral images by exploiting spectral-spatial information of superpixel via multiple kernels

Leyuan Fang, Shutao Li, Wuhui Duan, Jinchang Ren, Jon Atli Benediktsson

Research output: Contribution to journalArticle

141 Citations (Scopus)

Abstract

For the classification of hyperspectral images (HSIs), this paper presents a novel framework to effectively utilize the spectral-spatial information of superpixels via multiple kernels, termed as superpixel-based classification via multiple kernels (SC-MK). In HSI, each superpixel can be regarded as a shape-adaptive region which consists of a number of spatial-neighboring pixels with very similar spectral characteristics. Firstly, the proposed SC-MK method adopts an over-segmentation algorithm to cluster the HSI into many superpixels. Then, three kernels are separately employed for the utilization of the spectral information as well as spatial information within and among superpixels. Finally, the three kernels are combined together and incorporated into a support vector machines classifier. Experimental results on three widely used real HSIs indicate that the proposed SC-MK approach outperforms several well-known classification methods.
LanguageEnglish
Number of pages12
JournalIEEE Transactions on Geoscience and Remote Sensing
Early online date2 Jul 2015
DOIs
Publication statusPublished - 2015

Fingerprint

segmentation
Support vector machines
pixel
Classifiers
Pixels
method
support vector machine

Keywords

  • hyperspectral image
  • superpixel
  • multiple kernels
  • spectral-spatial image classification
  • support vector machines

Cite this

@article{fee9e9e6bb894a56a12b04a3966bc185,
title = "Classification of hyperspectral images by exploiting spectral-spatial information of superpixel via multiple kernels",
abstract = "For the classification of hyperspectral images (HSIs), this paper presents a novel framework to effectively utilize the spectral-spatial information of superpixels via multiple kernels, termed as superpixel-based classification via multiple kernels (SC-MK). In HSI, each superpixel can be regarded as a shape-adaptive region which consists of a number of spatial-neighboring pixels with very similar spectral characteristics. Firstly, the proposed SC-MK method adopts an over-segmentation algorithm to cluster the HSI into many superpixels. Then, three kernels are separately employed for the utilization of the spectral information as well as spatial information within and among superpixels. Finally, the three kernels are combined together and incorporated into a support vector machines classifier. Experimental results on three widely used real HSIs indicate that the proposed SC-MK approach outperforms several well-known classification methods.",
keywords = "hyperspectral image, superpixel, multiple kernels, spectral-spatial image classification, support vector machines",
author = "Leyuan Fang and Shutao Li and Wuhui Duan and Jinchang Ren and Benediktsson, {Jon Atli}",
note = "(c) 2015 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other users, 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 components of this work in other works.",
year = "2015",
doi = "10.1109/TGRS.2015.2445767",
language = "English",
journal = "IEEE Transactions on Geoscience and Remote Sensing",
issn = "0196-2892",
publisher = "IEEE",

}

Classification of hyperspectral images by exploiting spectral-spatial information of superpixel via multiple kernels. / Fang, Leyuan; Li, Shutao; Duan, Wuhui; Ren, Jinchang; Benediktsson, Jon Atli.

In: IEEE Transactions on Geoscience and Remote Sensing, 2015.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Classification of hyperspectral images by exploiting spectral-spatial information of superpixel via multiple kernels

AU - Fang, Leyuan

AU - Li, Shutao

AU - Duan, Wuhui

AU - Ren, Jinchang

AU - Benediktsson, Jon Atli

N1 - (c) 2015 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other users, 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 components of this work in other works.

PY - 2015

Y1 - 2015

N2 - For the classification of hyperspectral images (HSIs), this paper presents a novel framework to effectively utilize the spectral-spatial information of superpixels via multiple kernels, termed as superpixel-based classification via multiple kernels (SC-MK). In HSI, each superpixel can be regarded as a shape-adaptive region which consists of a number of spatial-neighboring pixels with very similar spectral characteristics. Firstly, the proposed SC-MK method adopts an over-segmentation algorithm to cluster the HSI into many superpixels. Then, three kernels are separately employed for the utilization of the spectral information as well as spatial information within and among superpixels. Finally, the three kernels are combined together and incorporated into a support vector machines classifier. Experimental results on three widely used real HSIs indicate that the proposed SC-MK approach outperforms several well-known classification methods.

AB - For the classification of hyperspectral images (HSIs), this paper presents a novel framework to effectively utilize the spectral-spatial information of superpixels via multiple kernels, termed as superpixel-based classification via multiple kernels (SC-MK). In HSI, each superpixel can be regarded as a shape-adaptive region which consists of a number of spatial-neighboring pixels with very similar spectral characteristics. Firstly, the proposed SC-MK method adopts an over-segmentation algorithm to cluster the HSI into many superpixels. Then, three kernels are separately employed for the utilization of the spectral information as well as spatial information within and among superpixels. Finally, the three kernels are combined together and incorporated into a support vector machines classifier. Experimental results on three widely used real HSIs indicate that the proposed SC-MK approach outperforms several well-known classification methods.

KW - hyperspectral image

KW - superpixel

KW - multiple kernels

KW - spectral-spatial image classification

KW - support vector machines

U2 - 10.1109/TGRS.2015.2445767

DO - 10.1109/TGRS.2015.2445767

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 -