An Advanced Facial Expression Recognition System Using Novel Machine Learning Approach by Comparing CNN over LeNet

Abstract

Aim:The primary goal of this study is to compare Convolutional Neural Networks (CNN) and LeNet for classification of facial
emotion recognition with improved accuracy. Materials and Methods: The data of the facial expressions is taken from the FER2013 available on kaggle. The Convolutional neural network (CNN) is the algorithm most popular way of analyzing images with best accuracy and LeNet is also an algorithm used here to compare the accuracy of Novel image classification. Results: The Convolution neural network (CNN) produces 82.14% accuracy in predicting facial expressions on the dataset,whereas LeNet produces 77.54% accuracy. Convolutional neural network (CNN) is better than LeNet. With (p<0.05), there is a statistically significant difference between the research groups. Conclusion: Convolutional Neural Network provides better outcomes in accuracy rate when compared to LeNet for predicting facial expressions.