Multimodal Biometric System Utilizing Palm Vein and Finger Vein Recognition

Abstract

Biometric authentication is gaining popularity due to its high security and ease of use compared to traditional methods. These technologies rely on unique individual traits, such as subcutaneous veins. The technology offers greater resistance to forgery (due to the difficulty of accessing subcutaneous veins) and environmental conditions, and is used in various fields where subcutaneous veins are only visible under near-infrared light. Our goal is to increase individual recognition accuracy using the veins of both right and left palms and the left index finger, employing feature extraction techniques. This involves starting with the extraction of the Region of Interest (ROI), selecting the best part of the image for subsequent feature extraction. Our proposal includes using feature extraction techniques with GABOR and LPQ filters. Finally, features are classified using KNN algorithms with various distance metrics (Euclidean, Cosine, City Block, Mahalanobis). Our study demonstrated highly encouraging performance, achieving a recognition rate of 100%.