Study Comparison Deep Learning and Support Vector Machine for Face Mask Detection

Abstract

Deep Learning (DL) and Support Vector Machine (SVM) was used for a plethora number of researches lately. Deep Learning works by representing data in layers of learning layers so that the representation becomes more meaningful, and Support Vector Machine tries to find the hyperplane that maximizes the margin between the hyperplane and the closest data points from each class so that the classification becomes more accurate. Both algorithms have proven to be powerful tools for any classification problem specially to classify or identify image patterns. However, the performance of machine learning algorithms can be affected by any factor, thus sometimes we found several algorithms that are generally known to be powerful, even showing unsatisfactory results. The purpose of this study is to compare the ability of classification methods Deep Learning and Support Vector Machine to detect face mask. Face mask detection has gained significant attention and importance in the context of public health and safety, particularly during the COVID-19 pandemic. The study revealed that Deep Learning algorithm performed better than the Support Vector Machine Algorithm and showed excellent performance in all four metrics. In particular, the Deep Learning algorithm achieved an average Sensitivity/Recall rate of 92%, a Specificity rate of 95.44%, a Precision rate of 95.28%, and an Accuracy rate of 93.72%.