Effective feature extraction and data reduction with hyperspectral imaging in remote sensing

Research output: Contribution to journalSpecial issue

66 Citations (Scopus)

Abstract

Although PCA has been widely used for feature extraction and data reduction, it suffers from three main drawbacks: high computational cost, large memory requirement and low efficacy in processing large datasets such as HSI. This column analysed two variations of PCA, namely SPCA and Seg-PCA. Seg-PCA can further improve classification accuracy whilst significantly reducing the computational cost and memory requirement, without requiring prior knowledge. There is potential to apply similar feature extraction and data reduction techniques in application areas beyond HSI when analysis of large dimensional datasets is required such as magnetic resonance imaging (MRI) and digital video processing.
LanguageEnglish
Pages149-154
Number of pages6
JournalIEEE Signal Processing Magazine
Volume31
Issue number4
DOIs
Publication statusPublished - Jul 2014

Fingerprint

Hyperspectral Imaging
Data Reduction
Remote Sensing
Feature Extraction
Feature extraction
Computational Cost
Remote sensing
Data reduction
Data storage equipment
Video Processing
Digital Video
Magnetic Resonance Imaging
Requirements
Magnetic resonance
Processing
Prior Knowledge
Large Data Sets
Efficacy
Costs
Imaging techniques

Keywords

  • hyperspectral Imaging
  • feature extraction
  • data reduction
  • remote sensing
  • covariance matrices
  • hypercubes
  • memory management
  • principal component analysis

Cite this

@article{da0adaa07c3146bc918c54db46eefb06,
title = "Effective feature extraction and data reduction with hyperspectral imaging in remote sensing",
abstract = "Although PCA has been widely used for feature extraction and data reduction, it suffers from three main drawbacks: high computational cost, large memory requirement and low efficacy in processing large datasets such as HSI. This column analysed two variations of PCA, namely SPCA and Seg-PCA. Seg-PCA can further improve classification accuracy whilst significantly reducing the computational cost and memory requirement, without requiring prior knowledge. There is potential to apply similar feature extraction and data reduction techniques in application areas beyond HSI when analysis of large dimensional datasets is required such as magnetic resonance imaging (MRI) and digital video processing.",
keywords = "hyperspectral Imaging, feature extraction, data reduction, remote sensing, covariance matrices, hypercubes, memory management, principal component analysis",
author = "Jinchang Ren and Jaime Zabalza and Stephen Marshall and Jiangbin Zheng",
year = "2014",
month = "7",
doi = "10.1109/MSP.2014.2312071",
language = "English",
volume = "31",
pages = "149--154",
journal = "IEEE Signal Processing Magazine",
issn = "1053-5888",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "4",

}

Effective feature extraction and data reduction with hyperspectral imaging in remote sensing. / Ren, Jinchang; Zabalza, Jaime; Marshall, Stephen; Zheng, Jiangbin.

In: IEEE Signal Processing Magazine, Vol. 31, No. 4, 07.2014, p. 149-154.

Research output: Contribution to journalSpecial issue

TY - JOUR

T1 - Effective feature extraction and data reduction with hyperspectral imaging in remote sensing

AU - Ren, Jinchang

AU - Zabalza, Jaime

AU - Marshall, Stephen

AU - Zheng, Jiangbin

PY - 2014/7

Y1 - 2014/7

N2 - Although PCA has been widely used for feature extraction and data reduction, it suffers from three main drawbacks: high computational cost, large memory requirement and low efficacy in processing large datasets such as HSI. This column analysed two variations of PCA, namely SPCA and Seg-PCA. Seg-PCA can further improve classification accuracy whilst significantly reducing the computational cost and memory requirement, without requiring prior knowledge. There is potential to apply similar feature extraction and data reduction techniques in application areas beyond HSI when analysis of large dimensional datasets is required such as magnetic resonance imaging (MRI) and digital video processing.

AB - Although PCA has been widely used for feature extraction and data reduction, it suffers from three main drawbacks: high computational cost, large memory requirement and low efficacy in processing large datasets such as HSI. This column analysed two variations of PCA, namely SPCA and Seg-PCA. Seg-PCA can further improve classification accuracy whilst significantly reducing the computational cost and memory requirement, without requiring prior knowledge. There is potential to apply similar feature extraction and data reduction techniques in application areas beyond HSI when analysis of large dimensional datasets is required such as magnetic resonance imaging (MRI) and digital video processing.

KW - hyperspectral Imaging

KW - feature extraction

KW - data reduction

KW - remote sensing

KW - covariance matrices

KW - hypercubes

KW - memory management

KW - principal component analysis

UR - http://ieeexplore.ieee.org/xpl/RecentIssue.jsp?punumber=79

U2 - 10.1109/MSP.2014.2312071

DO - 10.1109/MSP.2014.2312071

M3 - Special issue

VL - 31

SP - 149

EP - 154

JO - IEEE Signal Processing Magazine

T2 - IEEE Signal Processing Magazine

JF - IEEE Signal Processing Magazine

SN - 1053-5888

IS - 4

ER -