PHYSIOLOGICAL EMOTION RECOGNITION BY MEANS OF MACHINE LEARNING DEMONSTRATION

Abstract

The objective of Emotion Recognition is to identify the emotions of humans which can be captured either from face or from verbal communication. In this project, we focus on identifying human emotions from facial expressions. Facial expressions are important when we communicate, this will enable us to develop and decipher impressions of people around us. We make use of Image Processing Techniques such as Computer Vision Technique to recognize faces in both images and video streams and also Convolution Neural Networks from Deep Learning Technique has been used for better results and also to classify the image followed by feature extraction from the image. Using the extracted feature, the system is trained to classify a facial expression as happy, sad, disguised, neutral, pain or angry. This system works upon the human face for as we know that the face reflects the human brain activities or emotions. Hence, we impart intelligence to the computer in recognizing the human emotions by which we can easily identify the emotion of other humans without much effort.