Identification of Biometric Images by Machine Learning
Mariya Nazarkevych, Yaroslav Voznyi, Volodymyr Hrytsyk, Ivanna Klyujnyk, Bohdana Havrysh, Nataliia Lotoshynska
Abstract
The method of biometric prints classification by means of machine learning is offered. The “k-means” method was used to identify biometric images. Labeled sample data for learning and testing processes are generated. Experimental results point to the benefit of the presented method of integration of global and structured data and indicate that “k-means” is a promising approach to fingerprint classification. The development of biometrics leads to the creation of security systems with better degrees of recognition and with fewer errors than security systems on traditional media. Machine learning was performed using a number of samples from a known biometric database, and verification / testing was performed with samples from the same database that were not included in the training data set.