Skip to main navigation Skip to search Skip to main content

Spectral-spatial hyperspectral image classification via SVM and superpixel segmentation

  • Harbin Institute of Technology

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Integration of spatial information has recently emerged as a powerful tool in improving the classification accuracy of hyperspectral image (HSI). However, partitioning homogeneous regions of the HSI remains a challenging task. This paper proposes a novel spectral-spatial classification method inspired by the support vector machine (SVM) and superpixel segmentation. Core ideas of the proposed method are twofold: 1) the HSI is first classified by the pixel-wise classifier (i.e. SVM); 2) a fast superpixel segmentation-based spatial processing is, for the first time, introduced in this study to refine the homogeneity and consistency of the classification maps. Experiments are conducted on two benchmark HSIs (i.e. the Indian Pines data and the Washington, D.C. Mall data) with different spectral and spatial resolutions. It is found that the proposed method yields more accurate classification results compared to the state-of-the-Art techniques.

Original languageEnglish
Title of host publication2014 IEEE International Instrumentation and Measurement Technology Conference
Subtitle of host publicationInstrumentation and Measurement for Sustainable Development, I2MTC 2014 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages422-427
Number of pages6
ISBN (Print)9781467363853
DOIs
StatePublished - 2014
Event2014 IEEE International Instrumentation and Measurement Technology Conference: Instrumentation and Measurement for Sustainable Development, I2MTC 2014 - Montevideo, Uruguay
Duration: 12 May 201415 May 2014

Publication series

NameConference Record - IEEE Instrumentation and Measurement Technology Conference
ISSN (Print)1091-5281

Conference

Conference2014 IEEE International Instrumentation and Measurement Technology Conference: Instrumentation and Measurement for Sustainable Development, I2MTC 2014
Country/TerritoryUruguay
CityMontevideo
Period12/05/1415/05/14

Keywords

  • classification
  • entropy
  • graph
  • hyperspectral image (HSI)
  • superpixel segmentation
  • support vector machine (SVM)

Fingerprint

Dive into the research topics of 'Spectral-spatial hyperspectral image classification via SVM and superpixel segmentation'. Together they form a unique fingerprint.

Cite this