A Fuzzy Expert System for Facial Expression Recognition

Masoumeh Rezaei, Mansoureh Rezaei


This paper presents a method for facial expression recognition using fuzzy expert system. The proposed expert system consists of two main steps: First, the pre-processing part, the feature extraction step provides sufficient information for the inference engine. For this reason, NMF is used to preserve the representation of the original image. Additionally, it guarantees that both of the resulting low-dimensional basis and its accompanying weights are non-negative. Second, it allows for creating rules with the SGERD algorithm and inferencing them. The second step applies a suitable set of fuzzy rules and aggregates them towards the final decision. We applied our approach to the Japanese Female Facial Expression dataset for recognizing the facial expression states. Experimental results demonstrate superiority of the proposed approach to the compared methods in terms of classification rate


Nonnegative Matrix Factorization; Fuzzy Expert System; SGERD algorithm

Full Text:



  • There are currently no refbacks.

Creative Commons License
This work is licensed under a Creative Commons Attribution 3.0 License.

ISSN: 2180-1843

eISSN: 2289-8131