Institutional Repository of Vilnius University
Veidų atpažinimo algoritmų palyginimas. NT algoritmo realizacija /
Abstract
dc:descriptionFacial recognition has been one of the most interesting and popular areas of research in the last few years. These studies are being conducted on the need for automatic identification, the need for surveillance systems, and the design of the human-computer interface. Many articles have been published in the scientific literature that examine different factors such as: lighting, exposure, scale, or posture. The main goal of these studies was to achieve better recognition accuracy and detection speed, but it has not yet been studied whether different image recognition algorithms work in the same way using the same data sets. In this work, we will review general algorithm ideas and recognition principles using local binary patterns, Eigenfaces, support vector machines, and neural network algorithms. This paper analyzes the speed and accuracy of algorithms using different data sets, presents the disadvantages and advantages of all algorithms. Also, Using the most appropriate algorithm, a biometric facial recognition model is developed and used to create a security system as a visualization for a real-time personal authentication model.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2020
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Kvartūnas, Gražvydas,
- Contributors dc:contributor
-
- Bareikis, Gintautas
Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- lit
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD81804934&prefLang=en_US