Facial Emotion Recognition Analysis using Deep Convolution Neural Network

Abstract

Human emotions are the unpredictable fluctuations of the visual model which plays a vital role in non verbal communication. Emotions are the mental states of feelings that occur spontaneously and when combined with facial muscles form an expression. The basic emotions are happy, sad, angry, fear, disgust, and surprise. The paper presents a method for emotion recognition. The proposed methodology involves 2 stages, the first stage is image preprocessing which includes face detection using viola-jones algorithm .In the next stage detection of emotions is done by using convolution neural network(CNN) which shows the intensity changes from low level to high level of emotions on a face. when compared with existing system, our proposed system has better accuracy in recognition, also works as real time system with much less complexity than the existing system, In this system FERC-2013 dataset is used as standard image dataset. The assessment of proposed system gives a quick good result and provides an encouragement for the future researchers in emotion recognition system.