On the fly biometric identification system using hand-geometry

Classification (of information); Palmprint recognition

Fingerprint and face recognition systems, have become a norm due to mobile technologies, but its application in general settings, such as automated border gates using ePassports or at entrances to buildings or rooms are becoming popular as well. Regrettably, where current biometry scanning technologies are reliably suitable for single user devices, when hundreds or thousands of people need to be processed, shortcomings emerge. The requirement to physically stop during scanning, poses a limit to the number of people processed per a unit of time. Additionally, as majority of these systems require physical contact, hygienic concerns need to be considered, especially now in the aftermath of the COVID-19 pandemics. In this paper, we attempt to offer a solution to the issue of limited throughput and hygienic concerns, by developing a biometric system that allows on the fly identification. Using experimental setup of our design, we demonstrate that hand related biometrics can be acquired on the fly. Furthermore, using random decision forest classifier and database of over 560 unique images we obtained using said device, we establish, that the system is a viable solution for on the fly biometric identification.