Performance Comparison of ResNet50V2 and VGG16 Models for Feature Extraction in Deep Learning

With the prevalence of face masks due to the covid-19 pandemic, facial recognition technology commonly used for security screening in the workplace is facing significant challenges. These systems were typically trained on images of human faces without masks, making it difficult to identify individuals who are wearing masks. To address this issue, this chapter presents a deep learning-based model that is specifically designed to accurately recognize masked persons. The model is based on the ResNet-50 architecture and identifies individuals with face masks well. The research aims to compare the performance of two popular deep learning models, ResNet50V2 and VGG16, for feature extraction in image classification tasks. The study utilized a large dataset and measured the convergence rate and accuracy of both models. The results indicated that ResNet50V2 outperformed VGG16 in all aspects, exhibiting faster convergence and achieving higher accuracy.