Improved sparse representation using adaptive spatial support for effective target detection in hyperspectral imagery

Chunhui Zhao, Xiaohui Li, Jinchang Ren, Stephen Marshall

Research output: Contribution to journalArticle

38 Citations (Scopus)
240 Downloads (Pure)

Abstract

With increasing applications of hyperspectral imagery (HSI) in agriculture, mineralogy, military, and other fields, one of the fundamental tasks is accurate detection of the target of interest. In this article, improved sparse representation approaches using adaptive spatial support are proposed for effective target detection in HSI. For conventional sparse representation, an HSI pixel is represented as a sparse vector whose non-zero entries correspond to the weights of the selected training atoms from a structured dictionary. For improved sparse representation, spatial correlation and spectral similarity of adjacent neighbouring pixels are exploited as spatial support in this context. The size and shape of the spatial support is automatically determined using both adaptive window and adaptive neighbourhood strategies. Accordingly, a solution based on greedy pursuit algorithms is also given to solve the extended optimization problem in recovering the desired sparse representation. Comprehensive experiments on three different data sets using both visual inspection and quantitative evaluation are carried out. The results from these data sets have indicated that the proposed approaches help to generate improved results in terms of efficacy and efficiency.
Original languageEnglish
Pages (from-to)8669-8684
Number of pages16
JournalInternational Journal of Remote Sensing
Volume34
Issue number24
Early online date22 Nov 2013
DOIs
Publication statusPublished - 2013

Fingerprint

Target tracking
imagery
Pixels
Mineralogy
pixel
Glossaries
Agriculture
Inspection
Atoms
mineralogy
agriculture
Experiments
detection
experiment

Keywords

  • sparse representation
  • hyperspectral imagery
  • target detection
  • adaptive spatial support
  • remote sensing

Cite this

@article{f835466513c744918a96104fedeb66e0,
title = "Improved sparse representation using adaptive spatial support for effective target detection in hyperspectral imagery",
abstract = "With increasing applications of hyperspectral imagery (HSI) in agriculture, mineralogy, military, and other fields, one of the fundamental tasks is accurate detection of the target of interest. In this article, improved sparse representation approaches using adaptive spatial support are proposed for effective target detection in HSI. For conventional sparse representation, an HSI pixel is represented as a sparse vector whose non-zero entries correspond to the weights of the selected training atoms from a structured dictionary. For improved sparse representation, spatial correlation and spectral similarity of adjacent neighbouring pixels are exploited as spatial support in this context. The size and shape of the spatial support is automatically determined using both adaptive window and adaptive neighbourhood strategies. Accordingly, a solution based on greedy pursuit algorithms is also given to solve the extended optimization problem in recovering the desired sparse representation. Comprehensive experiments on three different data sets using both visual inspection and quantitative evaluation are carried out. The results from these data sets have indicated that the proposed approaches help to generate improved results in terms of efficacy and efficiency.",
keywords = "sparse representation, hyperspectral imagery, target detection, adaptive spatial support, remote sensing",
author = "Chunhui Zhao and Xiaohui Li and Jinchang Ren and Stephen Marshall",
year = "2013",
doi = "10.1080/01431161.2013.845924",
language = "English",
volume = "34",
pages = "8669--8684",
journal = "International Journal of Remote Sensing",
issn = "0143-1161",
number = "24",

}

TY - JOUR

T1 - Improved sparse representation using adaptive spatial support for effective target detection in hyperspectral imagery

AU - Zhao, Chunhui

AU - Li, Xiaohui

AU - Ren, Jinchang

AU - Marshall, Stephen

PY - 2013

Y1 - 2013

N2 - With increasing applications of hyperspectral imagery (HSI) in agriculture, mineralogy, military, and other fields, one of the fundamental tasks is accurate detection of the target of interest. In this article, improved sparse representation approaches using adaptive spatial support are proposed for effective target detection in HSI. For conventional sparse representation, an HSI pixel is represented as a sparse vector whose non-zero entries correspond to the weights of the selected training atoms from a structured dictionary. For improved sparse representation, spatial correlation and spectral similarity of adjacent neighbouring pixels are exploited as spatial support in this context. The size and shape of the spatial support is automatically determined using both adaptive window and adaptive neighbourhood strategies. Accordingly, a solution based on greedy pursuit algorithms is also given to solve the extended optimization problem in recovering the desired sparse representation. Comprehensive experiments on three different data sets using both visual inspection and quantitative evaluation are carried out. The results from these data sets have indicated that the proposed approaches help to generate improved results in terms of efficacy and efficiency.

AB - With increasing applications of hyperspectral imagery (HSI) in agriculture, mineralogy, military, and other fields, one of the fundamental tasks is accurate detection of the target of interest. In this article, improved sparse representation approaches using adaptive spatial support are proposed for effective target detection in HSI. For conventional sparse representation, an HSI pixel is represented as a sparse vector whose non-zero entries correspond to the weights of the selected training atoms from a structured dictionary. For improved sparse representation, spatial correlation and spectral similarity of adjacent neighbouring pixels are exploited as spatial support in this context. The size and shape of the spatial support is automatically determined using both adaptive window and adaptive neighbourhood strategies. Accordingly, a solution based on greedy pursuit algorithms is also given to solve the extended optimization problem in recovering the desired sparse representation. Comprehensive experiments on three different data sets using both visual inspection and quantitative evaluation are carried out. The results from these data sets have indicated that the proposed approaches help to generate improved results in terms of efficacy and efficiency.

KW - sparse representation

KW - hyperspectral imagery

KW - target detection

KW - adaptive spatial support

KW - remote sensing

UR - http://www.tandfonline.com/loi/tres20

U2 - 10.1080/01431161.2013.845924

DO - 10.1080/01431161.2013.845924

M3 - Article

VL - 34

SP - 8669

EP - 8684

JO - International Journal of Remote Sensing

JF - International Journal of Remote Sensing

SN - 0143-1161

IS - 24

ER -