Facial Expressions Recognition Using Markov Stationary Feature - Vector Quantization and Support Vector Machine Method

Abstract

Facial expression is a form of nonverbal communication that can convey the emotional state of someone to the person who observes it. Research on the recognition of facial expressions is one of the interesting fields in computer science. This research aims to improve the accuracy of recognition performance. The process carried out in this research is to perform feature extraction and image classification. Markov Stationary Feature - Vector Quantization (MSF-VQ) method is used for feature extraction and Support Vector Machine (SVM) for image classification. Data set in used is 1440 data with six classifications of facial expressions. The results of the testing showed 97.41% which stated that this method could be recommended to be applied in the facial expressions recognition.