Optimizing Attendance System: Integrating Liveness Detection and Deep Learning for Reliable Face Recognition

Abstract

The study focuses on using vitality detection and deep learning technologies in the context of facial recognition in an IT presence management project. The combination of deep learning with vitality detection provides a considerable advancement in security and effectiveness. This work integrated vitality-detecting technology with in-depth learning in facial recognition systems. Vitality detection technologies are used to verify the authenticity of persons by examining live indicators such as movements or facial expressions before face recognition. Meanwhile, deep learning is used to analyze and process facial photos correctly by learning from large amounts of data and recognizing facial features in depth. The study data set consists of 1300 photographs of professional school instructors taken with official authority. Model testing and training are carried out in the Google Colab environment, using Python and the Hardy package. The test findings showed an 87% accuracy in face recognition, proving the system's capacity to consistently identify persons and distinguish real from false ones. Furthermore, the performance of Liveness Detection achieves 92% accuracy, as does the integration of Live Detection technology with Deep Learning at 78%.

Keywords

• liveness detection, deep learning, face recognition, attendance management, integration technology