Hyperspectral data processing in a high performance computing environment

A parallel best band selection algorithm

Stefan Robila, Gerald Busardo

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

3 Citations (Scopus)

Abstract

Hyperspectral data are characterized by a richness of information unique among various visual representations of a scene by representing the information in a collection of grayscale images with each image corresponding to a narrow interval in the electromagnetic spectrum. Such detail allows for precise identification of materials in the scene and promises to support advances in imaging beyond the visible range. However, hyperspectral data are considerably large and cumbersome to process and efficient computing solutions based on high performance computing are needed. In this paper we first provide an overview of hyperspectral data and the current state of the art in the use of HPC for its processing. Next we discuss the concept of best band selection, a fundamental feature extraction problem in hyperspectral imagery that, besides exhaustive search has only non optimal solutions. We provide an elegant algorithm that performs an exhaustive search for the solution using a distributed, multicore environment and MPI in order to show how using such a solution provides significant improvement over traditional sequential platforms. Additional experiments on the robustness of the algorithm in terms of data and job sizes are also provided.

Original languageEnglish
Title of host publication2011 IEEE International Symposium on Parallel and Distributed Processing, Workshops and Phd Forum, IPDPSW 2011
Pages1424-1431
Number of pages8
DOIs
StatePublished - 20 Dec 2011
Event25th IEEE International Parallel and Distributed Processing Symposium, Workshops and Phd Forum, IPDPSW 2011 - Anchorage, AK, United States
Duration: 16 May 201120 May 2011

Other

Other25th IEEE International Parallel and Distributed Processing Symposium, Workshops and Phd Forum, IPDPSW 2011
CountryUnited States
CityAnchorage, AK
Period16/05/1120/05/11

Fingerprint

Hyperspectral Data
High Performance
Exhaustive Search
Computing
Hyperspectral Imagery
Feature Extraction
Feature extraction
Optimal Solution
Imaging
Robustness
Imaging techniques
Interval
Processing
Range of data
Experiment
Experiments

Keywords

  • Component selection
  • Distributed environments
  • Feature selection
  • Hyperspectral images
  • MPI

Cite this

Robila, S., & Busardo, G. (2011). Hyperspectral data processing in a high performance computing environment: A parallel best band selection algorithm. In 2011 IEEE International Symposium on Parallel and Distributed Processing, Workshops and Phd Forum, IPDPSW 2011 (pp. 1424-1431). [6008997] https://doi.org/10.1109/IPDPS.2011.282
Robila, Stefan ; Busardo, Gerald. / Hyperspectral data processing in a high performance computing environment : A parallel best band selection algorithm. 2011 IEEE International Symposium on Parallel and Distributed Processing, Workshops and Phd Forum, IPDPSW 2011. 2011. pp. 1424-1431
@inproceedings{af54e24621ac4f5aa763b64fc73ea18b,
title = "Hyperspectral data processing in a high performance computing environment: A parallel best band selection algorithm",
abstract = "Hyperspectral data are characterized by a richness of information unique among various visual representations of a scene by representing the information in a collection of grayscale images with each image corresponding to a narrow interval in the electromagnetic spectrum. Such detail allows for precise identification of materials in the scene and promises to support advances in imaging beyond the visible range. However, hyperspectral data are considerably large and cumbersome to process and efficient computing solutions based on high performance computing are needed. In this paper we first provide an overview of hyperspectral data and the current state of the art in the use of HPC for its processing. Next we discuss the concept of best band selection, a fundamental feature extraction problem in hyperspectral imagery that, besides exhaustive search has only non optimal solutions. We provide an elegant algorithm that performs an exhaustive search for the solution using a distributed, multicore environment and MPI in order to show how using such a solution provides significant improvement over traditional sequential platforms. Additional experiments on the robustness of the algorithm in terms of data and job sizes are also provided.",
keywords = "Component selection, Distributed environments, Feature selection, Hyperspectral images, MPI",
author = "Stefan Robila and Gerald Busardo",
year = "2011",
month = "12",
day = "20",
doi = "10.1109/IPDPS.2011.282",
language = "English",
isbn = "9780769543857",
pages = "1424--1431",
booktitle = "2011 IEEE International Symposium on Parallel and Distributed Processing, Workshops and Phd Forum, IPDPSW 2011",

}

Robila, S & Busardo, G 2011, Hyperspectral data processing in a high performance computing environment: A parallel best band selection algorithm. in 2011 IEEE International Symposium on Parallel and Distributed Processing, Workshops and Phd Forum, IPDPSW 2011., 6008997, pp. 1424-1431, 25th IEEE International Parallel and Distributed Processing Symposium, Workshops and Phd Forum, IPDPSW 2011, Anchorage, AK, United States, 16/05/11. https://doi.org/10.1109/IPDPS.2011.282

Hyperspectral data processing in a high performance computing environment : A parallel best band selection algorithm. / Robila, Stefan; Busardo, Gerald.

2011 IEEE International Symposium on Parallel and Distributed Processing, Workshops and Phd Forum, IPDPSW 2011. 2011. p. 1424-1431 6008997.

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

TY - GEN

T1 - Hyperspectral data processing in a high performance computing environment

T2 - A parallel best band selection algorithm

AU - Robila, Stefan

AU - Busardo, Gerald

PY - 2011/12/20

Y1 - 2011/12/20

N2 - Hyperspectral data are characterized by a richness of information unique among various visual representations of a scene by representing the information in a collection of grayscale images with each image corresponding to a narrow interval in the electromagnetic spectrum. Such detail allows for precise identification of materials in the scene and promises to support advances in imaging beyond the visible range. However, hyperspectral data are considerably large and cumbersome to process and efficient computing solutions based on high performance computing are needed. In this paper we first provide an overview of hyperspectral data and the current state of the art in the use of HPC for its processing. Next we discuss the concept of best band selection, a fundamental feature extraction problem in hyperspectral imagery that, besides exhaustive search has only non optimal solutions. We provide an elegant algorithm that performs an exhaustive search for the solution using a distributed, multicore environment and MPI in order to show how using such a solution provides significant improvement over traditional sequential platforms. Additional experiments on the robustness of the algorithm in terms of data and job sizes are also provided.

AB - Hyperspectral data are characterized by a richness of information unique among various visual representations of a scene by representing the information in a collection of grayscale images with each image corresponding to a narrow interval in the electromagnetic spectrum. Such detail allows for precise identification of materials in the scene and promises to support advances in imaging beyond the visible range. However, hyperspectral data are considerably large and cumbersome to process and efficient computing solutions based on high performance computing are needed. In this paper we first provide an overview of hyperspectral data and the current state of the art in the use of HPC for its processing. Next we discuss the concept of best band selection, a fundamental feature extraction problem in hyperspectral imagery that, besides exhaustive search has only non optimal solutions. We provide an elegant algorithm that performs an exhaustive search for the solution using a distributed, multicore environment and MPI in order to show how using such a solution provides significant improvement over traditional sequential platforms. Additional experiments on the robustness of the algorithm in terms of data and job sizes are also provided.

KW - Component selection

KW - Distributed environments

KW - Feature selection

KW - Hyperspectral images

KW - MPI

UR - http://www.scopus.com/inward/record.url?scp=83455262191&partnerID=8YFLogxK

U2 - 10.1109/IPDPS.2011.282

DO - 10.1109/IPDPS.2011.282

M3 - Conference contribution

SN - 9780769543857

SP - 1424

EP - 1431

BT - 2011 IEEE International Symposium on Parallel and Distributed Processing, Workshops and Phd Forum, IPDPSW 2011

ER -

Robila S, Busardo G. Hyperspectral data processing in a high performance computing environment: A parallel best band selection algorithm. In 2011 IEEE International Symposium on Parallel and Distributed Processing, Workshops and Phd Forum, IPDPSW 2011. 2011. p. 1424-1431. 6008997 https://doi.org/10.1109/IPDPS.2011.282