An Approach to Identify and Categorize Various Human Emotions Using Support Vector Machine

Abstract

Emotion recognition is very important in the upcoming security devices. This paper proposes a method to identify the emotions from human facial features expressions. The machine learning classifier is used to identify the emotions from facial expression is Support Vector Machine. Guassian radial basis function is used with the input images to categorize the emotion. The percentage of emotion is identified and categorization is done. Local binary pattern is used for texture classification of human face and to obtain the feature histogram. Matlab2017b version is used in implementing the paper using the machine learning technique. The accuracy obtained from the implemented method is 98%.

Keyword : Emotion recognition, Support vector machine, Guassian radial basis function, Local binary pattern, Machine learning.