A comparison of performance of crisp logic and probabilistic neural network for facial expression recognition

Automatic facial expression analysis is one of the most active research areas these days, as it has got various applications in human computer interaction systems. In order to analyse these expressions we need a system that takes up an image that contains a face and extract the features from it and classify it into one of the prototypes expressions. In this paper, a framework is designed that recognizes the facial expression by taking the input image from JAFFE database as well as from the web cam. The features are extracted using HAAR cascade and the classification is done using two techniques i.e. crisp logic and probabilistic logic. A table of results of both the algorithms is given at the end of the paper along with the bar graph that shows the comparison of their respective performances.