Identify Facial Micro Expression Using Support Vector Machine Compared with Artificial Neural Network to Improve Recall Parameter

Abstract

Aim: The creation of facial recognition systems, which are crucial in the modern world, is the aim of this research. The Novel Support Vector Machine and Artificial Neural Network are used in this study to build a rapid facial recognition technique in Python. In this research article, we compare the performance of Novel Support Vector Machine and Artificial Neural Network in facial recognition. Materials and Methods: The detailed studies of suggested algorithms are reviewed. The testing was carried out using a publicly accessible face database. Each algorithm is put to the test using ten different photos, each with a varied face expression and lighting. For the SPSS study, nearly 10 samples were taken to evaluate, compare, and understand the accuracy of proposed algorithms. For accuracy prediction, a G power of 80% is used in the SPSS software. The parameters considered are CI and alpha, which were determined as 0.003 (p < 0.05). For SVM group 1, 10 samples are taken and for Artificial Neural Network algorithm group 2, 10 samples are taken to compare the Recall for facial expressions. Result: The results of combining several feature extraction methods and classifiers were given and examined. SVM was shown to have the best accuracy, with a score of 97.01 % Conclusion: Emotion recognition is a promising technique for improving the efficiency of current image-based recognition techniques. In this research project.