PERANCANGAN JARINGAN SYARAF TIRUAN PERAMBATAN BALIK UNTUK AUTENTIKASI BIOMETRIK POLA PEMBULUH DARAH PUNGGUNG TANGAN BERBASIS TEKSTUR GLRLM

Digital security has become a critical aspect in the current era of information technology. Traditional identification methods such as passwords and PINs are starting to show limitations in overcoming increasingly complex security challenges. This research produces a system for identifying dorsal hand blood vessels as a biometric using feature extraction Gray Level Run Length Matrix (GLRLM) and the accurate Back Propagation Neural Network identification method which is expected to produce high accuracy compared to previous related research. This classification process has several stages, namely input image, image preprocessing, segmentation, feature extraction, image classification, and blood vessel image classification results. In this research, the largest test accuracy results were obtained in the fourth scenario, which was 87%. The test results showed that the percentage of success based on the five test scenarios had an average value of 82.52%. The average sensitivity based on five test scenarios was found to be 100% for Person 1, Person 2 100%, Person 31.66% and Person 4 98.35%. The average specification based on five test scenarios was obtained at 76.15% for Person 1, 100% for Person 2, 100% for Person 3 and for Person 4 100%. Judging from the results of the percentage of accuracy, sensitivity and specifications, it can be said that the system created is quite good in biometric identification using the object pattern of the back of the hand's blood vessels.

Keywords: Biometric, Digital Security, Dorsal Veins, GLRLM, Identification.