FPGA-based online learning hardware architecture for kernel fuzzy c-means algorithm

Chien Min Ou, Wen-Jyi Hwang, Ssu Min Yang

Research output: Contribution to journalArticle

Abstract

This paper presents a novel embedded system for the online training of kernel fuzzy c-means (KFCM) algorithm. A hardware architecture capable of accelerating the KFCM training process is proposed. The architecture is used as a coprocessor in the embedded system. It consists of efficient circuits for the computation of kernel functions, membership coefficients and cluster centers. In addition, the usual iterative operations for updating the membership matrix and cluster centers are merged into one single updating process to evade the large storage requirement. Experimental results show that the proposed solution is an effective alternative for image segmentation with low computational cost and low segmentation error rate.

Original languageEnglish
Pages (from-to)225-231
Number of pages7
JournalIndian Journal of Engineering and Materials Sciences
Volume20
Issue number3
Publication statusPublished - 2013 Jun 1

Fingerprint

Embedded systems
Field programmable gate arrays (FPGA)
Hardware
Membership functions
Image segmentation
Networks (circuits)
Costs
Coprocessor

Keywords

  • FPGA
  • Fuzzy clustering
  • Image segmentation
  • Reconfigurable computing
  • System-on-chip

ASJC Scopus subject areas

  • Materials Science(all)
  • Engineering(all)

Cite this

FPGA-based online learning hardware architecture for kernel fuzzy c-means algorithm. / Ou, Chien Min; Hwang, Wen-Jyi; Yang, Ssu Min.

In: Indian Journal of Engineering and Materials Sciences, Vol. 20, No. 3, 01.06.2013, p. 225-231.

Research output: Contribution to journalArticle

@article{c704156e55ad4f14b6fa2e7a438170d0,
title = "FPGA-based online learning hardware architecture for kernel fuzzy c-means algorithm",
abstract = "This paper presents a novel embedded system for the online training of kernel fuzzy c-means (KFCM) algorithm. A hardware architecture capable of accelerating the KFCM training process is proposed. The architecture is used as a coprocessor in the embedded system. It consists of efficient circuits for the computation of kernel functions, membership coefficients and cluster centers. In addition, the usual iterative operations for updating the membership matrix and cluster centers are merged into one single updating process to evade the large storage requirement. Experimental results show that the proposed solution is an effective alternative for image segmentation with low computational cost and low segmentation error rate.",
keywords = "FPGA, Fuzzy clustering, Image segmentation, Reconfigurable computing, System-on-chip",
author = "Ou, {Chien Min} and Wen-Jyi Hwang and Yang, {Ssu Min}",
year = "2013",
month = "6",
day = "1",
language = "English",
volume = "20",
pages = "225--231",
journal = "Indian Journal of Engineering and Materials Sciences",
issn = "0971-4588",
publisher = "National Institute of Science Communication and Information Resources (NISCAIR)",
number = "3",

}

TY - JOUR

T1 - FPGA-based online learning hardware architecture for kernel fuzzy c-means algorithm

AU - Ou, Chien Min

AU - Hwang, Wen-Jyi

AU - Yang, Ssu Min

PY - 2013/6/1

Y1 - 2013/6/1

N2 - This paper presents a novel embedded system for the online training of kernel fuzzy c-means (KFCM) algorithm. A hardware architecture capable of accelerating the KFCM training process is proposed. The architecture is used as a coprocessor in the embedded system. It consists of efficient circuits for the computation of kernel functions, membership coefficients and cluster centers. In addition, the usual iterative operations for updating the membership matrix and cluster centers are merged into one single updating process to evade the large storage requirement. Experimental results show that the proposed solution is an effective alternative for image segmentation with low computational cost and low segmentation error rate.

AB - This paper presents a novel embedded system for the online training of kernel fuzzy c-means (KFCM) algorithm. A hardware architecture capable of accelerating the KFCM training process is proposed. The architecture is used as a coprocessor in the embedded system. It consists of efficient circuits for the computation of kernel functions, membership coefficients and cluster centers. In addition, the usual iterative operations for updating the membership matrix and cluster centers are merged into one single updating process to evade the large storage requirement. Experimental results show that the proposed solution is an effective alternative for image segmentation with low computational cost and low segmentation error rate.

KW - FPGA

KW - Fuzzy clustering

KW - Image segmentation

KW - Reconfigurable computing

KW - System-on-chip

UR - http://www.scopus.com/inward/record.url?scp=84879038358&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84879038358&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:84879038358

VL - 20

SP - 225

EP - 231

JO - Indian Journal of Engineering and Materials Sciences

JF - Indian Journal of Engineering and Materials Sciences

SN - 0971-4588

IS - 3

ER -