Evolutionary structure optimization of hierarchical neural network for image recognition

Satoru Suzuki, Yasue Mitsukura

Research output: Contribution to journalArticle

Abstract

The purpose of this paper is to optimize the structure of hierarchical neural network (NN). In our proposed method, the structure optimization is considered as combinatorial optimization problem, and unnecessary connections in trained NN are eliminated by using genetic algorithm (GA).We focus on the NN which specialized for image recognition problems. In order to validate the usefulness of the proposed method, face recognition and texture classification examples are used. From the experimental results, it was shown that compact neural network was generated, keeping generalization performance by proposed method.

Original languageEnglish
Pages (from-to)983-989
Number of pages7
JournalIEEJ Transactions on Electronics, Information and Systems
Volume131
Issue number5
DOIs
Publication statusPublished - 2011
Externally publishedYes

Fingerprint

Image recognition
Neural networks
Combinatorial optimization
Face recognition
Textures
Genetic algorithms

Keywords

  • Face Recognition
  • Genetic Algorithm
  • Neural Network
  • Texture Classification

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Cite this

Evolutionary structure optimization of hierarchical neural network for image recognition. / Suzuki, Satoru; Mitsukura, Yasue.

In: IEEJ Transactions on Electronics, Information and Systems, Vol. 131, No. 5, 2011, p. 983-989.

Research output: Contribution to journalArticle

@article{6ee7e0067db24c62bbc7ab4a72048cbd,
title = "Evolutionary structure optimization of hierarchical neural network for image recognition",
abstract = "The purpose of this paper is to optimize the structure of hierarchical neural network (NN). In our proposed method, the structure optimization is considered as combinatorial optimization problem, and unnecessary connections in trained NN are eliminated by using genetic algorithm (GA).We focus on the NN which specialized for image recognition problems. In order to validate the usefulness of the proposed method, face recognition and texture classification examples are used. From the experimental results, it was shown that compact neural network was generated, keeping generalization performance by proposed method.",
keywords = "Face Recognition, Genetic Algorithm, Neural Network, Texture Classification",
author = "Satoru Suzuki and Yasue Mitsukura",
year = "2011",
doi = "10.1541/ieejeiss.131.983",
language = "English",
volume = "131",
pages = "983--989",
journal = "IEEJ Transactions on Electronics, Information and Systems",
issn = "0385-4221",
publisher = "The Institute of Electrical Engineers of Japan",
number = "5",

}

TY - JOUR

T1 - Evolutionary structure optimization of hierarchical neural network for image recognition

AU - Suzuki, Satoru

AU - Mitsukura, Yasue

PY - 2011

Y1 - 2011

N2 - The purpose of this paper is to optimize the structure of hierarchical neural network (NN). In our proposed method, the structure optimization is considered as combinatorial optimization problem, and unnecessary connections in trained NN are eliminated by using genetic algorithm (GA).We focus on the NN which specialized for image recognition problems. In order to validate the usefulness of the proposed method, face recognition and texture classification examples are used. From the experimental results, it was shown that compact neural network was generated, keeping generalization performance by proposed method.

AB - The purpose of this paper is to optimize the structure of hierarchical neural network (NN). In our proposed method, the structure optimization is considered as combinatorial optimization problem, and unnecessary connections in trained NN are eliminated by using genetic algorithm (GA).We focus on the NN which specialized for image recognition problems. In order to validate the usefulness of the proposed method, face recognition and texture classification examples are used. From the experimental results, it was shown that compact neural network was generated, keeping generalization performance by proposed method.

KW - Face Recognition

KW - Genetic Algorithm

KW - Neural Network

KW - Texture Classification

UR - http://www.scopus.com/inward/record.url?scp=80052417459&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=80052417459&partnerID=8YFLogxK

U2 - 10.1541/ieejeiss.131.983

DO - 10.1541/ieejeiss.131.983

M3 - Article

VL - 131

SP - 983

EP - 989

JO - IEEJ Transactions on Electronics, Information and Systems

JF - IEEJ Transactions on Electronics, Information and Systems

SN - 0385-4221

IS - 5

ER -