Fuzzy Integral-based Neural Network Ensemble for Facial Expression Recognition

DOI

10.2991/cisia-15.2015.193How to use a DOI?

Keywords

neural network ensembles; ensemble learning; information fusion; fuzzy integral; facial expression recognition

Abstract

Neural network is widely used in pattern classification. But an unavoidable fact is that many complex problems cannot get ideal correct rate of classification with a single neural network. Neural network ensembles provides a serial of theories and methods to simply ensemble outputs of some low performance neural networks, and the ensemble output can get higher correct rate. This paper tries to use fuzzy integral to complete the information integration of member classifier. A more effective fuzzy density function using in fuzzy integral was also proposed in this paper. Facial Expression Recognition (FER) was used to validate the method. Many simulation experiments were done to analyze the validity of neural network and different parameters. The Experiment data also proved the new fuzzy density function is more effective.

Copyright

© 2015, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Z.Y Wang
AU  - N.F Xiao
PY  - 2015/06
DA  - 2015/06
TI  - Fuzzy Integral-based Neural Network Ensemble for Facial Expression Recognition
BT  - Proceedings of the International Conference on Computer Information Systems and Industrial Applications
PB  - Atlantis Press
SP  - 709
EP  - 712
SN  - 2352-538X
UR  - https://doi.org/10.2991/cisia-15.2015.193
DO  - 10.2991/cisia-15.2015.193
ID  - Wang2015/06
ER  -