Deakin University
Browse

File(s) under permanent embargo

A New Facial Expression Recognition Scheme Based on Parallel Double Channel Convolutional Neural Network

Version 2 2024-06-04, 14:49
Version 1 2020-04-06, 11:24
conference contribution
posted on 2024-06-04, 14:49 authored by DT Li, F Jiang, YB Qin
© 2020, Springer Nature Switzerland AG. The conventional deep convolutional neural networks in facial expression recognition are confronted with the training inefficiency due to many layered structure with a large number of parameters, in order to cope with this challenge, in this work, an improved convolutional neural network—with parallel double channels, termed as PDC-CNN, is proposed. Within this model, we can get two different sized feature maps for the input image, and the two different sized feature maps are combined for the final recognition and judgment. In addition, in order to prevent over-fitting, we replace the traditional RuLU activation function with the Maxout model in the fully connected layer to optimize the performance of the network. We have trained and tested the new model on JAFFE dataset. Experimental results show that the proposed method can achieve 83% recognition rate, in comparison with the linear SVM, AlexNet and LeNet-5, the recognition rate of this method is improved by 14%–28%.

History

Pagination

560-569

Location

Phuket, Thailand

Start date

2020-03-23

End date

2020-03-26

ISSN

0302-9743

eISSN

1611-3349

ISBN-13

9783030420574

Language

eng

Publication classification

E1.1 Full written paper - refereed

Editor/Contributor(s)

Nguyen N, Jearanaitanakij K, Selamat A, Trawiński B, Chittayasothorn S

Title of proceedings

ACIIDS 2020 : Intelligent information and database systems : 12th Asian Conference, ACIIDS 2020, Phuket, Thailand, March 23-26, 2020, Proceedings.

Event

Asian Conference on Intelligent Information and Database Systems (12th : 2020 : Phuket, Thailand)

Publisher

Springer

Place of publication

Cham, Switzerland

Series

Lecture Notes in Computer Science; 12034

Usage metrics

    Research Publications

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC