Facial expression recognition based on two-step feature histogram optimization

DOI

10.2991/icmemtc-16.2016.310How to use a DOI?

Keywords

Facial expression; feature histogram; wight; support vector machine.

Abstract

The feature histogram is made of different labels containing information about patterns on a pixel-level. This means that pixels of the same label may come from different parts of a face. There must be some errors between the actual feature histogram and the accurate feature histogram which can completely represent the information of a face image. In order to overcome this shortcoming, we propose a facial expression recognition approach based on two-step features histogram optimization. This method requires two steps. In the first step, features histogram based on local binary patterns, uniform local binary patterns and local gradient coding to be extracted. In the second step, a suitable weight to multiply with features histogram extracted in the first step. And these features are classified by the support vector machine. Experiments show that our approach can obtain a higher recognition rate and maintain the time efficiency.

Copyright

© 2016, 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  - Ling Gan
AU  - Sisi Si
PY  - 2016/04
DA  - 2016/04
TI  - Facial expression recognition based on two-step feature histogram optimization
BT  - Proceedings of the 2016 3rd International Conference on Materials Engineering, Manufacturing Technology and Control
PB  - Atlantis Press
SP  - 1627
EP  - 1632
SN  - 2352-5401
UR  - https://doi.org/10.2991/icmemtc-16.2016.310
DO  - 10.2991/icmemtc-16.2016.310
ID  - Gan2016/04
ER  -