Efficient hardware architecture based on generalized Hebbian algorithm for texture classification

Shiow Jyu Lin, Yi Tsan Hung, Wen Jyi Hwang

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

The objective of this paper is to present an efficient hardware architecture for generalized Hebbian algorithm (GHA). In the architecture, the principal component computation and weight vector updating of the GHA are operated in parallel, so that the throughput of the circuit can be significantly enhanced. In addition, the weight vector updating process is separated into a number of stages for lowering area costs and increasing computational speed. To show the effectiveness of the circuit, a texture classification system based on the proposed architecture is designed. It is embedded in a system-on-programmable-chip (SOPC) platform for physical performance measurement. Experimental results show that the proposed architecture is an efficient design for attaining both high speed performance and low area costs.

Original languageEnglish
Pages (from-to)3248-3256
Number of pages9
JournalNeurocomputing
Volume74
Issue number17
DOIs
Publication statusPublished - 2011 Oct 1

Fingerprint

Textures
Hardware
Weights and Measures
Costs and Cost Analysis
Networks (circuits)
Costs
Throughput

Keywords

  • Generalized hebbian algorithm
  • Principal component analysis
  • System-on-programmable-chip

ASJC Scopus subject areas

  • Computer Science Applications
  • Cognitive Neuroscience
  • Artificial Intelligence

Cite this

Efficient hardware architecture based on generalized Hebbian algorithm for texture classification. / Lin, Shiow Jyu; Hung, Yi Tsan; Hwang, Wen Jyi.

In: Neurocomputing, Vol. 74, No. 17, 01.10.2011, p. 3248-3256.

Research output: Contribution to journalArticle

@article{cef40c9c273a462fbafc8ba10a228966,
title = "Efficient hardware architecture based on generalized Hebbian algorithm for texture classification",
abstract = "The objective of this paper is to present an efficient hardware architecture for generalized Hebbian algorithm (GHA). In the architecture, the principal component computation and weight vector updating of the GHA are operated in parallel, so that the throughput of the circuit can be significantly enhanced. In addition, the weight vector updating process is separated into a number of stages for lowering area costs and increasing computational speed. To show the effectiveness of the circuit, a texture classification system based on the proposed architecture is designed. It is embedded in a system-on-programmable-chip (SOPC) platform for physical performance measurement. Experimental results show that the proposed architecture is an efficient design for attaining both high speed performance and low area costs.",
keywords = "Generalized hebbian algorithm, Principal component analysis, System-on-programmable-chip",
author = "Lin, {Shiow Jyu} and Hung, {Yi Tsan} and Hwang, {Wen Jyi}",
year = "2011",
month = "10",
day = "1",
doi = "10.1016/j.neucom.2011.05.010",
language = "English",
volume = "74",
pages = "3248--3256",
journal = "Neurocomputing",
issn = "0925-2312",
publisher = "Elsevier",
number = "17",

}

TY - JOUR

T1 - Efficient hardware architecture based on generalized Hebbian algorithm for texture classification

AU - Lin, Shiow Jyu

AU - Hung, Yi Tsan

AU - Hwang, Wen Jyi

PY - 2011/10/1

Y1 - 2011/10/1

N2 - The objective of this paper is to present an efficient hardware architecture for generalized Hebbian algorithm (GHA). In the architecture, the principal component computation and weight vector updating of the GHA are operated in parallel, so that the throughput of the circuit can be significantly enhanced. In addition, the weight vector updating process is separated into a number of stages for lowering area costs and increasing computational speed. To show the effectiveness of the circuit, a texture classification system based on the proposed architecture is designed. It is embedded in a system-on-programmable-chip (SOPC) platform for physical performance measurement. Experimental results show that the proposed architecture is an efficient design for attaining both high speed performance and low area costs.

AB - The objective of this paper is to present an efficient hardware architecture for generalized Hebbian algorithm (GHA). In the architecture, the principal component computation and weight vector updating of the GHA are operated in parallel, so that the throughput of the circuit can be significantly enhanced. In addition, the weight vector updating process is separated into a number of stages for lowering area costs and increasing computational speed. To show the effectiveness of the circuit, a texture classification system based on the proposed architecture is designed. It is embedded in a system-on-programmable-chip (SOPC) platform for physical performance measurement. Experimental results show that the proposed architecture is an efficient design for attaining both high speed performance and low area costs.

KW - Generalized hebbian algorithm

KW - Principal component analysis

KW - System-on-programmable-chip

UR - http://www.scopus.com/inward/record.url?scp=80052917466&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=80052917466&partnerID=8YFLogxK

U2 - 10.1016/j.neucom.2011.05.010

DO - 10.1016/j.neucom.2011.05.010

M3 - Article

AN - SCOPUS:80052917466

VL - 74

SP - 3248

EP - 3256

JO - Neurocomputing

JF - Neurocomputing

SN - 0925-2312

IS - 17

ER -