Target detection in hyperspectral images based on independent component analysis

Stefan Robila, Pramod K. Varshney

Research output: Contribution to journalArticle

25 Citations (Scopus)

Abstract

The paper presents an algorithm based on Independent Component Analysis (ICA) for the detection of small targets present in hyperspectral images. ICA is a multivariate data analysis method that attempts to produce statistically independent components. This method is based on fourth order statistics. Small, man-made targets in a natural background can be seen as anomalies in the image scene and correspond to independent components in the ICA model. The algorithm described here starts by preprocessing the hyperspectral data through centering and sphering, thus eliminating the first and second order statistics. It then separates the features present in the image using an ICA based algorithm. The method involves a gradient descent minimization of the mutual information between frames. The resulting frames are ranked according to their kurtosis (defined by normalized fourth order moment of the sample distribution). High kurtosis valued frames indicate the presence of small man-made targets. Thresholding the frames using zero detection in their histogram further identifies the targets. The effectiveness of the method has been studied on data from the hyperspectral digital imagery collection experiment (HYDICE). Preliminary results show that small targets present in the image are separated from the background in different frames and that information pertaining to them is concentrated in these frames. Frame selection using kurtosis and thresholding leads to automated identification of the targets. The experiments show that the method provides a promising new approach for target detection.

Original languageEnglish
Pages (from-to)173-182
Number of pages10
JournalProceedings of SPIE-The International Society for Optical Engineering
Volume4726
DOIs
StatePublished - 1 Jan 2002

Fingerprint

Hyperspectral Image
Target Detection
Independent component analysis
Independent Component Analysis
Target tracking
kurtosis
Target
Kurtosis
Statistics
Thresholding
Order Statistics
Fourth Order
statistics
Multivariate Data Analysis
Experiments
Hyperspectral Data
Gradient Descent
descent
preprocessing
Mutual Information

Keywords

  • Feature selection
  • Higher order statistics
  • Hyperspectral imagery
  • Independent component analysis
  • Mutual information based feature separation
  • Target detection

Cite this

@article{aa576a2f62e24d50a006ac8e95f61292,
title = "Target detection in hyperspectral images based on independent component analysis",
abstract = "The paper presents an algorithm based on Independent Component Analysis (ICA) for the detection of small targets present in hyperspectral images. ICA is a multivariate data analysis method that attempts to produce statistically independent components. This method is based on fourth order statistics. Small, man-made targets in a natural background can be seen as anomalies in the image scene and correspond to independent components in the ICA model. The algorithm described here starts by preprocessing the hyperspectral data through centering and sphering, thus eliminating the first and second order statistics. It then separates the features present in the image using an ICA based algorithm. The method involves a gradient descent minimization of the mutual information between frames. The resulting frames are ranked according to their kurtosis (defined by normalized fourth order moment of the sample distribution). High kurtosis valued frames indicate the presence of small man-made targets. Thresholding the frames using zero detection in their histogram further identifies the targets. The effectiveness of the method has been studied on data from the hyperspectral digital imagery collection experiment (HYDICE). Preliminary results show that small targets present in the image are separated from the background in different frames and that information pertaining to them is concentrated in these frames. Frame selection using kurtosis and thresholding leads to automated identification of the targets. The experiments show that the method provides a promising new approach for target detection.",
keywords = "Feature selection, Higher order statistics, Hyperspectral imagery, Independent component analysis, Mutual information based feature separation, Target detection",
author = "Stefan Robila and Varshney, {Pramod K.}",
year = "2002",
month = "1",
day = "1",
doi = "10.1117/12.477024",
language = "English",
volume = "4726",
pages = "173--182",
journal = "Proceedings of SPIE - The International Society for Optical Engineering",
issn = "0277-786X",
publisher = "SPIE",

}

Target detection in hyperspectral images based on independent component analysis. / Robila, Stefan; Varshney, Pramod K.

In: Proceedings of SPIE-The International Society for Optical Engineering, Vol. 4726, 01.01.2002, p. 173-182.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Target detection in hyperspectral images based on independent component analysis

AU - Robila, Stefan

AU - Varshney, Pramod K.

PY - 2002/1/1

Y1 - 2002/1/1

N2 - The paper presents an algorithm based on Independent Component Analysis (ICA) for the detection of small targets present in hyperspectral images. ICA is a multivariate data analysis method that attempts to produce statistically independent components. This method is based on fourth order statistics. Small, man-made targets in a natural background can be seen as anomalies in the image scene and correspond to independent components in the ICA model. The algorithm described here starts by preprocessing the hyperspectral data through centering and sphering, thus eliminating the first and second order statistics. It then separates the features present in the image using an ICA based algorithm. The method involves a gradient descent minimization of the mutual information between frames. The resulting frames are ranked according to their kurtosis (defined by normalized fourth order moment of the sample distribution). High kurtosis valued frames indicate the presence of small man-made targets. Thresholding the frames using zero detection in their histogram further identifies the targets. The effectiveness of the method has been studied on data from the hyperspectral digital imagery collection experiment (HYDICE). Preliminary results show that small targets present in the image are separated from the background in different frames and that information pertaining to them is concentrated in these frames. Frame selection using kurtosis and thresholding leads to automated identification of the targets. The experiments show that the method provides a promising new approach for target detection.

AB - The paper presents an algorithm based on Independent Component Analysis (ICA) for the detection of small targets present in hyperspectral images. ICA is a multivariate data analysis method that attempts to produce statistically independent components. This method is based on fourth order statistics. Small, man-made targets in a natural background can be seen as anomalies in the image scene and correspond to independent components in the ICA model. The algorithm described here starts by preprocessing the hyperspectral data through centering and sphering, thus eliminating the first and second order statistics. It then separates the features present in the image using an ICA based algorithm. The method involves a gradient descent minimization of the mutual information between frames. The resulting frames are ranked according to their kurtosis (defined by normalized fourth order moment of the sample distribution). High kurtosis valued frames indicate the presence of small man-made targets. Thresholding the frames using zero detection in their histogram further identifies the targets. The effectiveness of the method has been studied on data from the hyperspectral digital imagery collection experiment (HYDICE). Preliminary results show that small targets present in the image are separated from the background in different frames and that information pertaining to them is concentrated in these frames. Frame selection using kurtosis and thresholding leads to automated identification of the targets. The experiments show that the method provides a promising new approach for target detection.

KW - Feature selection

KW - Higher order statistics

KW - Hyperspectral imagery

KW - Independent component analysis

KW - Mutual information based feature separation

KW - Target detection

UR - http://www.scopus.com/inward/record.url?scp=0036031933&partnerID=8YFLogxK

U2 - 10.1117/12.477024

DO - 10.1117/12.477024

M3 - Article

AN - SCOPUS:0036031933

VL - 4726

SP - 173

EP - 182

JO - Proceedings of SPIE - The International Society for Optical Engineering

JF - Proceedings of SPIE - The International Society for Optical Engineering

SN - 0277-786X

ER -