A Novelty of Face Recognition using Hybrid Artificial Intelligence Method

In the context of the COVID-19 pandemic that occurred in the past year, people have had to adjust their daily lives to minimize physical contact, including the shift towards online learning, as traditional classroom settings became impractical. Educational institutions have heavily relied on online learning platforms, leading to challenges in measuring and maintaining students' engagement and attendance. Consequently, there has been a surge in research on facial recognition technology to aid in monitoring attendance and participation in online education. However, existing facial recognition systems often face limitations, particularly in terms of excessive data usage for training. The objective of this study is to enhance the efficiency of existing facial recognition systems and reduce resource consumption. In this research, we employed feature extraction techniques in conjunction with GoogLeNet Convolutional Neural Networks and Support Vector Machine, specifically focusing on extracting crucial facial features that remain consistent for training. The achieved facial recognition accuracy in this study reached up to 93%.