Facial expression recognition system with five layer convolution neural network

Abstract

Now a days facial expression recognition field is growing and playing important
role in communication. Facial expression recognition has been an active research area in
the past ten years, with growing application areas including avatar animation, neuro
marketing and sociable robots. The recognition of facial expressions is not an easy
problem for machine learning methods, since people can vary significantly in the way
they show their expressions. However, facial expressions change so subtly that
recognition accuracy of most largely depend on feature extraction. In this work, we
propose two independent methods for this very task. The first method uses facial
expression recognition system using with five layer convolutional neural network, while
the second method is an 8-layer convolutional neural network (CNN) or Alex net.In the
first method to achieve facial expression recognition based on a deep CNN. Firstly we
implement face detection by using Haar-like features and histogram equalization. Then we
construct a five-layer CNN architecture, including two convolutional layers and two
subsampling layers (C-S-C-S). Finally, a Softmax classifier is used for multi-
classification.Second method,created an 8-layer CNN with five convolutional layers, and
three fully connected layers. This module, to reduce redundancy of same features learned,
considers mutual information between filters of the same layer, and processes the best set
of features for the next layer.

MPhil CS 2017-2018 INT Dr. Elizabeth Sherly