A fast source separation algorithm for hyperspectral image processing

Stefan Robila, Pramod K. Varshney

Research output: Contribution to conferencePaperResearchpeer-review

15 Citations (Scopus)

Abstract

This paper describes a new algorithm for feature extraction in hyperspectral images based on Independent Component Analysis (ICA). The improvement introduced aims at reducing the computation times without decreasing the accuracy. Instead of using the entire image, we perform ICA processing on a subset of representative pixel vectors obtained through spectral screening. Spectral screening is a technique that measures the similarity between pixel vectors by calculating the angle between them. In multispectral/hyperspectral imagery, the independent components can be associated with features present in the image. ICA projects them in different image frames. The features are separated using an algorithm involving gradient descent minimization of the mutual information between frames. The effectiveness of the proposed algorithm (SSICA) has been tested by performing target detection on data from the hyperspectral digital imagery collection experiment (HYDICE). Small targets present in the image are separated from the background in different frames and the information pertaining to them is concentrated in these frames. Further selection using kurtosis, skewness and histogram thresholding lead to automated detection of the targets allowing a quantitative assessment of the results. When compared with a target detection ICA algorithm previously introduced by the authors, SSICA achieves similar accuracy, and, at the same time, considerable speedup is obtained.

Original languageEnglish
Pages3516-3518
Number of pages3
StatePublished - 1 Jan 2002
Event2002 IEEE International Geoscience and Remote Sensing Symposium (IGARSS 2002) - Toronto, Ont., Canada
Duration: 24 Jun 200228 Jun 2002

Other

Other2002 IEEE International Geoscience and Remote Sensing Symposium (IGARSS 2002)
CountryCanada
CityToronto, Ont.
Period24/06/0228/06/02

Fingerprint

Source separation
Independent component analysis
image processing
Image processing
Target tracking
Screening
Pixels
pixel
imagery
Set theory
skewness
Feature extraction
histogram
Processing
analysis
Experiments
detection
experiment

Cite this

Robila, S., & Varshney, P. K. (2002). A fast source separation algorithm for hyperspectral image processing. 3516-3518. Paper presented at 2002 IEEE International Geoscience and Remote Sensing Symposium (IGARSS 2002), Toronto, Ont., Canada.
Robila, Stefan ; Varshney, Pramod K. / A fast source separation algorithm for hyperspectral image processing. Paper presented at 2002 IEEE International Geoscience and Remote Sensing Symposium (IGARSS 2002), Toronto, Ont., Canada.3 p.
@conference{af4787a67ecb4275b0e86807732bcad7,
title = "A fast source separation algorithm for hyperspectral image processing",
abstract = "This paper describes a new algorithm for feature extraction in hyperspectral images based on Independent Component Analysis (ICA). The improvement introduced aims at reducing the computation times without decreasing the accuracy. Instead of using the entire image, we perform ICA processing on a subset of representative pixel vectors obtained through spectral screening. Spectral screening is a technique that measures the similarity between pixel vectors by calculating the angle between them. In multispectral/hyperspectral imagery, the independent components can be associated with features present in the image. ICA projects them in different image frames. The features are separated using an algorithm involving gradient descent minimization of the mutual information between frames. The effectiveness of the proposed algorithm (SSICA) has been tested by performing target detection on data from the hyperspectral digital imagery collection experiment (HYDICE). Small targets present in the image are separated from the background in different frames and the information pertaining to them is concentrated in these frames. Further selection using kurtosis, skewness and histogram thresholding lead to automated detection of the targets allowing a quantitative assessment of the results. When compared with a target detection ICA algorithm previously introduced by the authors, SSICA achieves similar accuracy, and, at the same time, considerable speedup is obtained.",
author = "Stefan Robila and Varshney, {Pramod K.}",
year = "2002",
month = "1",
day = "1",
language = "English",
pages = "3516--3518",
note = "null ; Conference date: 24-06-2002 Through 28-06-2002",

}

Robila, S & Varshney, PK 2002, 'A fast source separation algorithm for hyperspectral image processing' Paper presented at 2002 IEEE International Geoscience and Remote Sensing Symposium (IGARSS 2002), Toronto, Ont., Canada, 24/06/02 - 28/06/02, pp. 3516-3518.

A fast source separation algorithm for hyperspectral image processing. / Robila, Stefan; Varshney, Pramod K.

2002. 3516-3518 Paper presented at 2002 IEEE International Geoscience and Remote Sensing Symposium (IGARSS 2002), Toronto, Ont., Canada.

Research output: Contribution to conferencePaperResearchpeer-review

TY - CONF

T1 - A fast source separation algorithm for hyperspectral image processing

AU - Robila, Stefan

AU - Varshney, Pramod K.

PY - 2002/1/1

Y1 - 2002/1/1

N2 - This paper describes a new algorithm for feature extraction in hyperspectral images based on Independent Component Analysis (ICA). The improvement introduced aims at reducing the computation times without decreasing the accuracy. Instead of using the entire image, we perform ICA processing on a subset of representative pixel vectors obtained through spectral screening. Spectral screening is a technique that measures the similarity between pixel vectors by calculating the angle between them. In multispectral/hyperspectral imagery, the independent components can be associated with features present in the image. ICA projects them in different image frames. The features are separated using an algorithm involving gradient descent minimization of the mutual information between frames. The effectiveness of the proposed algorithm (SSICA) has been tested by performing target detection on data from the hyperspectral digital imagery collection experiment (HYDICE). Small targets present in the image are separated from the background in different frames and the information pertaining to them is concentrated in these frames. Further selection using kurtosis, skewness and histogram thresholding lead to automated detection of the targets allowing a quantitative assessment of the results. When compared with a target detection ICA algorithm previously introduced by the authors, SSICA achieves similar accuracy, and, at the same time, considerable speedup is obtained.

AB - This paper describes a new algorithm for feature extraction in hyperspectral images based on Independent Component Analysis (ICA). The improvement introduced aims at reducing the computation times without decreasing the accuracy. Instead of using the entire image, we perform ICA processing on a subset of representative pixel vectors obtained through spectral screening. Spectral screening is a technique that measures the similarity between pixel vectors by calculating the angle between them. In multispectral/hyperspectral imagery, the independent components can be associated with features present in the image. ICA projects them in different image frames. The features are separated using an algorithm involving gradient descent minimization of the mutual information between frames. The effectiveness of the proposed algorithm (SSICA) has been tested by performing target detection on data from the hyperspectral digital imagery collection experiment (HYDICE). Small targets present in the image are separated from the background in different frames and the information pertaining to them is concentrated in these frames. Further selection using kurtosis, skewness and histogram thresholding lead to automated detection of the targets allowing a quantitative assessment of the results. When compared with a target detection ICA algorithm previously introduced by the authors, SSICA achieves similar accuracy, and, at the same time, considerable speedup is obtained.

UR - http://www.scopus.com/inward/record.url?scp=0036398882&partnerID=8YFLogxK

M3 - Paper

SP - 3516

EP - 3518

ER -

Robila S, Varshney PK. A fast source separation algorithm for hyperspectral image processing. 2002. Paper presented at 2002 IEEE International Geoscience and Remote Sensing Symposium (IGARSS 2002), Toronto, Ont., Canada.