Class Attendance System Based-on Palm Vein as Biometric Information

Modern educational institutions demand efficient and accurate attendance tracking systems to streamline administrative processes and ensure the utilization of instructional time. This paper presents a novel approach to class attendance management through the utilization of palm vein recognition technology and Convolutional Neural Networks (CNNs) implemented in MATLAB. 

Palm vein recognition offers a secure and reliable biometric authentication method due to its unique vascular pattern, which is virtually impossible to duplicate. The proposed system extracts and processes palm vein images, utilizing near-infrared light to capture the vascular pattern beneath the skin's surface. These images are then preprocessed to enhance features and reduce noise, ensuring optimal input for subsequent analysis. The core of the system involves the integration of Convolutional Neural Networks, a deep learning architecture renowned for its prowess in feature extraction from visual data. 

The CNN model is trained on a diverse dataset of palm vein images, enabling it to learn discriminative patterns inherent to individuals' vascular structures. The trained CNN can accurately classify and identify individuals, allowing for efficient attendance management.

Keywords: Image processing, palm vein dataset, Deep learning, save the attendance tracking.