Facial expression identification system using fisher linear discriminant analysis and K-nearest neighbor methods

Abstract EN

Facial expression system has become an important and effective research area in many fields such as cognitive processes, medical care and interaction between man and computer.

A facial expressions recognition system using each of FLDA with K-nearest neighbors (K-NN) classifier is introduced in this research.

The system is applied to recognize various basic facial expressions such as happy, neutral, angry, disgust, sad, fear and surprise, in the Karolinska Directed Emotional Faces (KDEF) and Japanese Female Facial Expressions (JAFFE) database.

The experimental results on JAFFE database proved that the proposed method is robust with good accuracy compared to other approaches.

The accuracy rate of the system achieved 95.09% and 94% when the proposed method was tested.