Automatic face emotion recognition and classification using Genetic Algorithm

Abstract

Facial expressions analysis most significant part for human computer interaction.Now days, face emotion recognition is most important application of computer vision that can be used for security, entertainment and human machine interface.Automatic face emotion recognition is still challenging & emerging problem with many applications such an automatic surveillance, robot motion, video indexing and retrieval and monitoring systems.Emotion recognition and classification depends upon gesture, pose, facial expression, speech and behavioral reactions, etc.In this paper, an automatic emotion recognition and classification method is based on Genetic Algorithm and on neural network.This system consists of 3 steps which automatically detect the face emotion image: First, pre-processing such as adjusting contrast, colour segmentation, filtering, and edge detection is applied on the input image.Secondly, features are extracted with projection profile method due to high speed which has taken as processed input image.Finally, in third stage to compute optimized parameters of eyes and lip through the GA, then emotions (neutral, happy, sad, dislike, angry, surprise and fear) is classified using artificial neural network.The proposed system is tested on a face emotion image.The obtained results show that better performance of genetic algorithm along with neural network.