A SYSTEM AND METHOD FOR MONITORING FACIAL EMOTIONS AND BEHAVIOR USING DEEP NEURAL NETWORK

Abstract The present disclosure seeks to provide a system and method for monitoring facial emotions and behaviour using deep neural network. In this disclosure, the system comprises: a pre-processing unit for detecting face and thereby segmenting said detected face from input image/video; a feature extraction module for extracting features from said greyscale image/video; a classification module for detecting facial emotion and behavior of a person using machine learning technique; a decision unit for comparing and matching feature of detected facial emotion of a person from a dataset and a prediction module for predicting facial emotion and behavior from a set of facial emotions and behaviors. The method comprises: detecting face and thereby segmenting said detected face from input image/video; aligning and cropping segmented face from said input image/video; converting color image/video into greyscale and thereby generating eigen faces; extracting features from said greyscale image/video using a deep neural network; detecting facial emotion and behavior of a person using extracted features through machine learning technique; comparing and matching feature of detected facial emotion and behavior of a person form a dataset of a plurality of images/videos of facial emotions and behaviors; and predicting facial emotion and behavior of said person from a set of facial emotions and behaviors. b* o* t= 2i t"