Návrh biometrického systému pro klasifikaci otisků prstů

Abstract

The thesis focuses on the development of a simulation environment for the evaluation of artificial intelligence classification methods and different machine learning approaches (convolutional neural networks – GoogLeNet and ResNet-101) in the context of biometric fingerprint identification. The convolutional neural networks classification algorithm is tested on two datasets of different sizes. The validation accuracy is assessed based on the combination of the hyperparameter values  batch size, learning rate and number of epochs. The theoretical part of the thesis focuses on the definition of biometrics and biometric systems in the context of fingerprint-based security identification. The practical part of the thesis contains basic information on the use of neural networks to classify image data and its implantation on a specific fingerprint dataset.