The Successive Projection Algorithm (SPA), an Algorithm with a Spatial Constraint for the Automatic Search of Endmembers in Hyperspectral Data
Spectral mixing is a problem inherent to remote sensing data and results in fewimage pixel spectra representing "pure" targets. Linear spectral mixture analysis isdesigned to address this problem and it assumes that the pixel-to-pixel variability in ascene results from varying proportions...
Published in: | Sensors |
---|---|
Main Authors: | , , |
Format: | Text |
Language: | English |
Published: |
Molecular Diversity Preservation International
2008
|
Subjects: | |
Online Access: | https://doi.org/10.3390/s8021321 |
id |
ftmdpi:oai:mdpi.com:/1424-8220/8/2/1321/ |
---|---|
record_format |
openpolar |
spelling |
ftmdpi:oai:mdpi.com:/1424-8220/8/2/1321/ 2023-08-20T04:05:24+02:00 The Successive Projection Algorithm (SPA), an Algorithm with a Spatial Constraint for the Automatic Search of Endmembers in Hyperspectral Data Jinkai Zhang Benoit Rivard D. M. Rogge 2008-02-22 application/pdf https://doi.org/10.3390/s8021321 EN eng Molecular Diversity Preservation International Remote Sensors https://dx.doi.org/10.3390/s8021321 https://creativecommons.org/licenses/by/3.0/ Sensors; Volume 8; Issue 2; Pages: 1321-1342 hyperspectral spectral unmixing endmember simplex Text 2008 ftmdpi https://doi.org/10.3390/s8021321 2023-07-31T20:21:38Z Spectral mixing is a problem inherent to remote sensing data and results in fewimage pixel spectra representing "pure" targets. Linear spectral mixture analysis isdesigned to address this problem and it assumes that the pixel-to-pixel variability in ascene results from varying proportions of spectral endmembers. In this paper we present adifferent endmember-search algorithm called the Successive Projection Algorithm (SPA).SPA builds on convex geometry and orthogonal projection common to other endmembersearch algorithms by including a constraint on the spatial adjacency of endmembercandidate pixels. Consequently it can reduce the susceptibility to outlier pixels andgenerates realistic endmembers.This is demonstrated using two case studies (AVIRISCuprite cube and Probe-1 imagery for Baffin Island) where image endmembers can bevalidated with ground truth data. The SPA algorithm extracts endmembers fromhyperspectral data without having to reduce the data dimensionality. It uses the spectralangle (alike IEA) and the spatial adjacency of pixels in the image to constrain the selectionof candidate pixels representing an endmember. We designed SPA based on theobservation that many targets have spatial continuity (e.g. bedrock lithologies) in imageryand thus a spatial constraint would be beneficial in the endmember search. An additionalproduct of the SPA is data describing the change of the simplex volume ratio between successive iterations during the endmember extraction. It illustrates the influence of a newendmember on the data structure, and provides information on the convergence of thealgorithm. It can provide a general guideline to constrain the total number of endmembersin a search. Text Baffin Island Baffin MDPI Open Access Publishing Baffin Island Sensors 8 2 1321 1342 |
institution |
Open Polar |
collection |
MDPI Open Access Publishing |
op_collection_id |
ftmdpi |
language |
English |
topic |
hyperspectral spectral unmixing endmember simplex |
spellingShingle |
hyperspectral spectral unmixing endmember simplex Jinkai Zhang Benoit Rivard D. M. Rogge The Successive Projection Algorithm (SPA), an Algorithm with a Spatial Constraint for the Automatic Search of Endmembers in Hyperspectral Data |
topic_facet |
hyperspectral spectral unmixing endmember simplex |
description |
Spectral mixing is a problem inherent to remote sensing data and results in fewimage pixel spectra representing "pure" targets. Linear spectral mixture analysis isdesigned to address this problem and it assumes that the pixel-to-pixel variability in ascene results from varying proportions of spectral endmembers. In this paper we present adifferent endmember-search algorithm called the Successive Projection Algorithm (SPA).SPA builds on convex geometry and orthogonal projection common to other endmembersearch algorithms by including a constraint on the spatial adjacency of endmembercandidate pixels. Consequently it can reduce the susceptibility to outlier pixels andgenerates realistic endmembers.This is demonstrated using two case studies (AVIRISCuprite cube and Probe-1 imagery for Baffin Island) where image endmembers can bevalidated with ground truth data. The SPA algorithm extracts endmembers fromhyperspectral data without having to reduce the data dimensionality. It uses the spectralangle (alike IEA) and the spatial adjacency of pixels in the image to constrain the selectionof candidate pixels representing an endmember. We designed SPA based on theobservation that many targets have spatial continuity (e.g. bedrock lithologies) in imageryand thus a spatial constraint would be beneficial in the endmember search. An additionalproduct of the SPA is data describing the change of the simplex volume ratio between successive iterations during the endmember extraction. It illustrates the influence of a newendmember on the data structure, and provides information on the convergence of thealgorithm. It can provide a general guideline to constrain the total number of endmembersin a search. |
format |
Text |
author |
Jinkai Zhang Benoit Rivard D. M. Rogge |
author_facet |
Jinkai Zhang Benoit Rivard D. M. Rogge |
author_sort |
Jinkai Zhang |
title |
The Successive Projection Algorithm (SPA), an Algorithm with a Spatial Constraint for the Automatic Search of Endmembers in Hyperspectral Data |
title_short |
The Successive Projection Algorithm (SPA), an Algorithm with a Spatial Constraint for the Automatic Search of Endmembers in Hyperspectral Data |
title_full |
The Successive Projection Algorithm (SPA), an Algorithm with a Spatial Constraint for the Automatic Search of Endmembers in Hyperspectral Data |
title_fullStr |
The Successive Projection Algorithm (SPA), an Algorithm with a Spatial Constraint for the Automatic Search of Endmembers in Hyperspectral Data |
title_full_unstemmed |
The Successive Projection Algorithm (SPA), an Algorithm with a Spatial Constraint for the Automatic Search of Endmembers in Hyperspectral Data |
title_sort |
successive projection algorithm (spa), an algorithm with a spatial constraint for the automatic search of endmembers in hyperspectral data |
publisher |
Molecular Diversity Preservation International |
publishDate |
2008 |
url |
https://doi.org/10.3390/s8021321 |
geographic |
Baffin Island |
geographic_facet |
Baffin Island |
genre |
Baffin Island Baffin |
genre_facet |
Baffin Island Baffin |
op_source |
Sensors; Volume 8; Issue 2; Pages: 1321-1342 |
op_relation |
Remote Sensors https://dx.doi.org/10.3390/s8021321 |
op_rights |
https://creativecommons.org/licenses/by/3.0/ |
op_doi |
https://doi.org/10.3390/s8021321 |
container_title |
Sensors |
container_volume |
8 |
container_issue |
2 |
container_start_page |
1321 |
op_container_end_page |
1342 |
_version_ |
1774715918608236544 |