Evolutionary-fuzzy-integral-based convolutional neural networks for facial image classification

Cheng Jian Lin, Chun Hui Lin, Chi Chia Sun, Shyh Hau Wang

Research output: Contribution to journalArticle

Abstract

Various optimization methods and network architectures are used by convolutional neural networks (CNNs). Each optimization method and network architecture style have their own advantages and representation abilities. To make the most of these advantages, evolutionary-fuzzyintegral- based convolutional neural networks (EFI-CNNs) are proposed in this paper. The proposed EFI-CNNs were verified by way of face classification of age and gender. The trained CNNs’ outputs were set as inputs of a fuzzy integral. The classification results were operated using either Sugeno or Choquet output rules. The conventional fuzzy density values of the fuzzy integral were decided by heuristic experiments. In this paper, particle swarm optimization (PSO) was used to adaptively find optimal fuzzy density values. To combine the advantages of each CNN type, the evaluation of each CNN type in EFI-CNNs is necessary. Three CNN structures, AlexNet, very deep convolutional neural network (VGG16), and GoogLeNet, and three databases, computational intelligence application laboratory (CIA), Morph, and cross-age celebrity dataset (CACD2000), were used in experiments to classify age and gender. The experimental results show that the proposed method achieved 5.95% and 3.1% higher accuracy, respectively, in classifying age and gender.

Original languageEnglish
Article number997
JournalElectronics (Switzerland)
Volume8
Issue number9
DOIs
Publication statusPublished - 2019 Sep

Fingerprint

Image classification
Neural networks
Network architecture
Particle swarm optimization (PSO)
Artificial intelligence
Experiments

All Science Journal Classification (ASJC) codes

  • Control and Systems Engineering
  • Signal Processing
  • Hardware and Architecture
  • Computer Networks and Communications
  • Electrical and Electronic Engineering

Cite this

@article{ff15c0e1afb749848621bc4c7bc2bb4d,
title = "Evolutionary-fuzzy-integral-based convolutional neural networks for facial image classification",
abstract = "Various optimization methods and network architectures are used by convolutional neural networks (CNNs). Each optimization method and network architecture style have their own advantages and representation abilities. To make the most of these advantages, evolutionary-fuzzyintegral- based convolutional neural networks (EFI-CNNs) are proposed in this paper. The proposed EFI-CNNs were verified by way of face classification of age and gender. The trained CNNs’ outputs were set as inputs of a fuzzy integral. The classification results were operated using either Sugeno or Choquet output rules. The conventional fuzzy density values of the fuzzy integral were decided by heuristic experiments. In this paper, particle swarm optimization (PSO) was used to adaptively find optimal fuzzy density values. To combine the advantages of each CNN type, the evaluation of each CNN type in EFI-CNNs is necessary. Three CNN structures, AlexNet, very deep convolutional neural network (VGG16), and GoogLeNet, and three databases, computational intelligence application laboratory (CIA), Morph, and cross-age celebrity dataset (CACD2000), were used in experiments to classify age and gender. The experimental results show that the proposed method achieved 5.95{\%} and 3.1{\%} higher accuracy, respectively, in classifying age and gender.",
author = "Lin, {Cheng Jian} and Lin, {Chun Hui} and Sun, {Chi Chia} and Wang, {Shyh Hau}",
year = "2019",
month = "9",
doi = "10.3390/electronics8090997",
language = "English",
volume = "8",
journal = "Electronics (Switzerland)",
issn = "2079-9292",
publisher = "Multidisciplinary Digital Publishing Institute (MDPI)",
number = "9",

}

Evolutionary-fuzzy-integral-based convolutional neural networks for facial image classification. / Lin, Cheng Jian; Lin, Chun Hui; Sun, Chi Chia; Wang, Shyh Hau.

In: Electronics (Switzerland), Vol. 8, No. 9, 997, 09.2019.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Evolutionary-fuzzy-integral-based convolutional neural networks for facial image classification

AU - Lin, Cheng Jian

AU - Lin, Chun Hui

AU - Sun, Chi Chia

AU - Wang, Shyh Hau

PY - 2019/9

Y1 - 2019/9

N2 - Various optimization methods and network architectures are used by convolutional neural networks (CNNs). Each optimization method and network architecture style have their own advantages and representation abilities. To make the most of these advantages, evolutionary-fuzzyintegral- based convolutional neural networks (EFI-CNNs) are proposed in this paper. The proposed EFI-CNNs were verified by way of face classification of age and gender. The trained CNNs’ outputs were set as inputs of a fuzzy integral. The classification results were operated using either Sugeno or Choquet output rules. The conventional fuzzy density values of the fuzzy integral were decided by heuristic experiments. In this paper, particle swarm optimization (PSO) was used to adaptively find optimal fuzzy density values. To combine the advantages of each CNN type, the evaluation of each CNN type in EFI-CNNs is necessary. Three CNN structures, AlexNet, very deep convolutional neural network (VGG16), and GoogLeNet, and three databases, computational intelligence application laboratory (CIA), Morph, and cross-age celebrity dataset (CACD2000), were used in experiments to classify age and gender. The experimental results show that the proposed method achieved 5.95% and 3.1% higher accuracy, respectively, in classifying age and gender.

AB - Various optimization methods and network architectures are used by convolutional neural networks (CNNs). Each optimization method and network architecture style have their own advantages and representation abilities. To make the most of these advantages, evolutionary-fuzzyintegral- based convolutional neural networks (EFI-CNNs) are proposed in this paper. The proposed EFI-CNNs were verified by way of face classification of age and gender. The trained CNNs’ outputs were set as inputs of a fuzzy integral. The classification results were operated using either Sugeno or Choquet output rules. The conventional fuzzy density values of the fuzzy integral were decided by heuristic experiments. In this paper, particle swarm optimization (PSO) was used to adaptively find optimal fuzzy density values. To combine the advantages of each CNN type, the evaluation of each CNN type in EFI-CNNs is necessary. Three CNN structures, AlexNet, very deep convolutional neural network (VGG16), and GoogLeNet, and three databases, computational intelligence application laboratory (CIA), Morph, and cross-age celebrity dataset (CACD2000), were used in experiments to classify age and gender. The experimental results show that the proposed method achieved 5.95% and 3.1% higher accuracy, respectively, in classifying age and gender.

UR - http://www.scopus.com/inward/record.url?scp=85073387132&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85073387132&partnerID=8YFLogxK

U2 - 10.3390/electronics8090997

DO - 10.3390/electronics8090997

M3 - Article

AN - SCOPUS:85073387132

VL - 8

JO - Electronics (Switzerland)

JF - Electronics (Switzerland)

SN - 2079-9292

IS - 9

M1 - 997

ER -