A Critical Review of Finger Vein Recognition Techniques for Human Identification

The specific parts of human body used for vein recognition are finger, palm, hand dorsal and wrist. The merit of the popularity of a finger vein recognition method depends upon key techniques, evaluation method, application in mobile and portable devices, ease of use, future trends, etc. Compared to other biometric behaviours like fingerprints, footprints, palmprints, face, retina, etc., the vein-based recognition is much secure and reliable as it is very difficult to forge. Several methods and techniques based on traditional computer vision and deep learning have been developed in last 10 years for finger vein recognition, however, a very limited use f finger vein is implemented in commercial applications for authentication. A comparative study of methods and techniques with their merits and demerits for finger vein recognition in last 10 years are presented in this paper. The study also presents the comparative accuracies of different methods/techniques and a platform for emerging approaches towards use of finger vein as biometric recognition.