Enhancing ATM Security: A Finger Vein Biometrics Approach

The purpose of this study is to highlight a revolutionary real-time embedded finger-vein detection system that was developed for the purpose of enhancing the safety of automated teller machines. The technology takes advantage of an embedded platform and integrates a CNN algorithm for the detection of finger veins. The picture capture module which takes datasets of couple of vein images for image processing, the embedded main board that uses for the display the output and sends the message to user. The embedded main board handles processing the output that are collected from the finger veins by the image acquisition module, which is responsible for collecting the photos. For the purpose of executing the recognition algorithm and communicating with peripheral devices, this board is equipped with a communication port, memory (flash), and a microcontroller chip. That makes it easier to display the results of recognition and to receive input from members of the user community. The system, which was designed as an intelligent security solution, offers remote monitoring through internet access and provides transaction verification through the use of GSM technology. The final results showed encouraging outcomes. The study yielded the following results: a mean value of 41.5751, a standard deviation of 51.2256, a mean square error of 9.9691e-0.4, a PSNR of 78.1443, and an entropy of 5.0595.