Hybrid MPI/OpenMP Implementation of PCA

Dalia Shouman Ibrahim; Salma Hamdy

doi:10.1109/INTELCIS.2017.8260048

Hybrid MPI/OpenMP Implementation of PCA

Dalia Shouman Ibrahim
, Salma Hamdy

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

2 Citations (Scopus)

371 Downloads (Pure)

Abstract

Most surveillance systems depend on fully automated face recognition applications. The main concern is achieving high accuracy in real time. Principle Component Analysis algorithm is used for reducing the number of variables and getting the maximum variance between low dimensional data. The proposed approaches focus on data partitioning to minimize the execution time of the algorithm by distributing data over a cluster with parallel computing architecture. The first approach achieves 2975X and 102X relatively faster than the sequential implementation in the training and recognition phases, respectively. However, the second approach achieves 74X relatively faster than the sequential implementation in the recognition phase.

Original language	English
Title of host publication	Proceedings of the 2017 IEEE Eighth International Conference on Intelligent Computing and Information Systems (ICICIS 2017)
Place of Publication	Cairo, Egypt
Publisher	IEEE
Pages	205-211
Number of pages	7
ISBN (Electronic)	9781538608210
ISBN (Print)	9781538608227
DOIs	https://doi.org/10.1109/INTELCIS.2017.8260048
Publication status	Published - 5 Dec 2017
Externally published	Yes
Event	8th International Conference on Intelligent Computing and Information Systems - Ain Shams University, Cairo, Egypt Duration: 5 Dec 2017 → 7 Dec 2017 Conference number: 8 http://net2.asu.edu.eg/icicis/2017/

Publication series

Name
ISSN (Print)	1687-1103

Conference

Conference	8th International Conference on Intelligent Computing and Information Systems
Abbreviated title	ICICIS 2017
Country/Territory	Egypt
City	Cairo
Period	5/12/17 → 7/12/17
Internet address	http://net2.asu.edu.eg/icicis/2017/

Keywords

Face recognition
PCA
Hybrid MPI/OpenMP
Parallel programming
Hybrid memory architecture

Access to Document

10.1109/INTELCIS.2017.8260048

ElSayed_Hybrid MPI-OpenMPImplementationOfPCA_Author_2017Accepted author manuscript, 1.02 MB

Salma Hamdy Elsayed
- s.elsayedabertay.acuk
- Department of Cybersecurity and Computing - Academic Disciplinary Officer, Lecturer
Person: Academic

Parallel architecture for face recognition using MPI
Ibrahim, D. S. & Hamdy, S., 17 Feb 2017, In: International Journal of Advanced Computer Science and Applications (IJACSA). 8, 1, p. 425-430 6 p., 54.
Research output: Contribution to journal › Article › peer-review

Open Access
File
129 Downloads (Pure)

Cite this

@inproceedings{335269ac8d4f408e82e352f0d4d3a651,

title = "Hybrid MPI/OpenMP Implementation of PCA",

abstract = "Most surveillance systems depend on fully automated face recognition applications. The main concern is achieving high accuracy in real time. Principle Component Analysis algorithm is used for reducing the number of variables and getting the maximum variance between low dimensional data. The proposed approaches focus on data partitioning to minimize the execution time of the algorithm by distributing data over a cluster with parallel computing architecture. The first approach achieves 2975X and 102X relatively faster than the sequential implementation in the training and recognition phases, respectively. However, the second approach achieves 74X relatively faster than the sequential implementation in the recognition phase.",

author = "Ibrahim, \{Dalia Shouman\} and Salma Hamdy",

year = "2017",

month = dec,

day = "5",

doi = "10.1109/INTELCIS.2017.8260048",

language = "English",

isbn = "9781538608227",

publisher = "IEEE ",

pages = "205--211",

booktitle = "Proceedings of the 2017 IEEE Eighth International Conference on Intelligent Computing and Information Systems (ICICIS 2017)",

note = "8th International Conference on Intelligent Computing and Information Systems, ICICIS 2017 ; Conference date: 05-12-2017 Through 07-12-2017",

url = "http://net2.asu.edu.eg/icicis/2017/",

}

Ibrahim, DS & Hamdy, S 2017, Hybrid MPI/OpenMP Implementation of PCA. in Proceedings of the 2017 IEEE Eighth International Conference on Intelligent Computing and Information Systems (ICICIS 2017). IEEE , Cairo, Egypt, pp. 205-211, 8th International Conference on Intelligent Computing and Information Systems, Cairo, Egypt, 5/12/17. https://doi.org/10.1109/INTELCIS.2017.8260048

TY - GEN

T1 - Hybrid MPI/OpenMP Implementation of PCA

AU - Ibrahim, Dalia Shouman

AU - Hamdy, Salma

N1 - Conference code: 8

PY - 2017/12/5

Y1 - 2017/12/5

N2 - Most surveillance systems depend on fully automated face recognition applications. The main concern is achieving high accuracy in real time. Principle Component Analysis algorithm is used for reducing the number of variables and getting the maximum variance between low dimensional data. The proposed approaches focus on data partitioning to minimize the execution time of the algorithm by distributing data over a cluster with parallel computing architecture. The first approach achieves 2975X and 102X relatively faster than the sequential implementation in the training and recognition phases, respectively. However, the second approach achieves 74X relatively faster than the sequential implementation in the recognition phase.

AB - Most surveillance systems depend on fully automated face recognition applications. The main concern is achieving high accuracy in real time. Principle Component Analysis algorithm is used for reducing the number of variables and getting the maximum variance between low dimensional data. The proposed approaches focus on data partitioning to minimize the execution time of the algorithm by distributing data over a cluster with parallel computing architecture. The first approach achieves 2975X and 102X relatively faster than the sequential implementation in the training and recognition phases, respectively. However, the second approach achieves 74X relatively faster than the sequential implementation in the recognition phase.

U2 - 10.1109/INTELCIS.2017.8260048

DO - 10.1109/INTELCIS.2017.8260048

M3 - Conference contribution

SN - 9781538608227

SP - 205

EP - 211

BT - Proceedings of the 2017 IEEE Eighth International Conference on Intelligent Computing and Information Systems (ICICIS 2017)

PB - IEEE

CY - Cairo, Egypt

T2 - 8th International Conference on Intelligent Computing and Information Systems

Y2 - 5 December 2017 through 7 December 2017

ER -

Hybrid MPI/OpenMP Implementation of PCA

Abstract

Publication series

Conference

Keywords

Access to Document

Fingerprint

Profiles

Salma Hamdy Elsayed

Research output

Parallel architecture for face recognition using MPI

Cite this