Dorsal Hand Veins Based Biometric Identification System Using Deep Learning

Identification systems have become biometric-based, especially with the increase in performance rates of machine learning methods. Biometric identification offer a high level security by using reliable, difficult-to-change parameters. In this thesis, biometric system is proposed dorsal hand vein patterns. The relevant has been tested on sample dataset literature. number data were increased adding noisy to set used. classification was made preprocessed images SVM, ANN, LDA + KNN, and CNN It determined that highest accuracy achieved when used, method provides higher compared other With system, after multiple runs, an average 99.64% method.