Enhanced K-Means Based Facial Expressions Recognition System
Automatic facial expression recognition is an interesting and active research topic for research in the recent years.
Facial expression recognition plays a vital role in Human Computer Interaction. Facial expressions are one of the key features
of facial recognition. In this research work, Enhanced K-Means algorithm is proposed for classification of facial expressions
from frontal facial images. To classify the expressions, algorithm uses two features: density of pixels and ratio of height to
width of cropped boundary regions. The recognition system comprises preprocessing, feature extraction and expression
classification. Based on the features extracted, Enhanced K-Means algorithm will classify the expressions into one of the
expressions happy, sad and neutral. Expression classification will apply on the dataset of 200 images of KDEF (Karolinska
Directed Emotional Face Database) database and expected to improve the performance of existing recognition system.