Dorsal Vein Biometric Prototype System using ResNet18

This study presents the development of a prototype device for dorsal vein biometric recognition using the ResNet18 architecture. The proposed system achieved high accuracy in recognizing the dorsal vein patterns from the right and left hand. The system was trained using the Stochastic Gradient Descent with Momentum (SGDM) optimizer algorithm, with batch learning used to update the model every 10 sets of data for a maximum of 6 epochs. A simulation study was conducted to evaluate the system, which achieved an average accuracy of 98.99% and 98.8% when using right and left dorsal images as features, respectively. The proposed system provides a non-invasive and user-friendly approach to biometric identification that can be utilized in various applications such as access control, authentication, and identification. The results demonstrate the promising potential of the ResNet18 architecture for dorsal vein biometric recognition, offering high accuracy and fast processing times. Future research can be conducted to enhance the system's performance by utilizing larger datasets, optimizing hyperparameters, and exploring alternative architectures.