|
|
Core of the Multimodal Biometric System
Pokorný, Karel ; Drahanský, Martin (referee) ; Váňa, Jan (advisor)
The aim of this thesis is a design and realization of the core of multimodal biometric system. First part of the thesis sumarizes contemporary knowledge about biometric systems and about combination of their outputs. Second part introduces concept and implementation of multimodal biometric system, which uses weighted score combination and user-specific weights.
|
|
|
Liveness Detection on Finger Using Various Light Sources
Vančo, Petr ; Kanich, Ondřej (referee) ; Drahanský, Martin (advisor)
The purpose of this work was to increase quality, security and reliability in biometric systems. This work presents electric and optical characteristics of a human finger. In the first part of this work you can find a theoretical basis about biometric systems, a skin and light. The next part is devoted to electric characteristics of a human finger. The third part of this work describes design, implementation and results of testing optical characteristics of a finger. In the end you can find information about some of the improvements for the future.
|
|
|
Brute Force Attack on Access System Using Synthetic Fingerprints
Polehňa, Dominik ; Goldmann, Tomáš (referee) ; Kanich, Ondřej (advisor)
The work deals with the resistance of the VeriFinger comparison algorithm to synthetic fingerprints. The basics of fingerprints, comparison algorithms and synthetic fingerprint generators are gradually analyzed. In total, three experiments were designed to test the resistance of the algorithm using synthetic fingerprints. An application using the VeriFinger algorithm was implemented to evaluate individual experiments. A total of 2 800 000 synthetic fingerprints were generated across all experiments. The first experiment showed that random fingerprint generation was not effective for real fingerprints, but was somewhat sufficient for synthetic fingerprints. The second experiment proved that comparison and the comparison score could be used to estimate the class of the identified fingerprint, and in the third experiment, by narrowing the generation to one class, it raised the total number of matched fingerprints. The results of individual experiments were written and their possible extension was suggested.
|
|
|
Analysis of Fingerprint Spoofs Made from Various Materials
Sláma, Vladimír ; Drahanský, Martin (referee) ; Kanich, Ondřej (advisor)
The goal of the work is production of fingerprint spoofs from various materials and develop- ment of algorithm, that analyzes them. Production of fingerprint spoofs is based on casting using printed circuit board (PCB) and other molds. For fingerprint spoofs these materials were used: Herkules, plasticine, Funny gummy, wax, Moody putty and liquid latex. For analysis of the differencies between spoof and the real fingerprint Harris detector was used and then the difference score is counted based on points of interests found. Results are compared with evaluation from VeriFinger software, that can measure matching score of two fingerprints. In most of the cases, the results have same trends when compared.
|
|
|
Support for Fingerprint Authentication
Bartoň, Jaroslav ; Beran, Vítězslav (referee) ; Mlích, Jozef (advisor)
The goal of the thesis is the finger-print authentication support within the Linux operating system and the K Desktop Environment (KDE). Theoretical part of the thesis firstly explains main IT security terms and ways to proof the identity. Secondly it describes biometric systems and types of processed biometric characteristics. Lastly the features of finger-prints, their markants as well as types of scanners used in scanning the finger-prints and ways to analyze the scanned material have been elaborated. Practical solution part of the thesis develops and establishes finger-print management application and plugin for KDM graphics login manager.
|
|
|
Liveness Detection on Touchless Fingerprint Scanner
Fořtová, Kateřina ; Kanich, Ondřej (referee) ; Heidari, Mona (advisor)
This Bachelor's Thesis is focused on liveness detection of fingerprints with using touchless sensor. Work summarizes theoretical introduction to biometrics, fingerprint processing and some of present researches for liveness detection. The new approach is introduced with using Local Binary Pattern algorithm, Sobel and Laplacian operator and Wavelet transform. Artificial Neural Networks, Support Vector Machines and Decision Trees were used for final classification. Several experiments with dataset illuminated by lights with various wavelengths were realized. It was discovered, that fingerprints illuminated by red light reached the best accuracy 90.1% compared to other considered wavelenghts of visible light. The classification with vector based on Local Binary Pattern achieved average accuracy 89.8%, accuracy with vector based on Sobel and Laplacian operator was 91.5%. Several Wavelet families were used for Wavelet transform during experiments. The best accuracy achieved wavelets of Biorthogonal spline wavelet family (85.1%) and wavelets from Reverse biorthogonal spline wavelet family (86.6%).
|
|
|
Extraction of Detailed Information from Fingerprint
Češka, Petr ; Sakin, Martin (referee) ; Kanich, Ondřej (advisor)
This bachelor's thesis deals with the extraction of detailed information from a fingerprint , especially width , direction and density of papillary lines . The main goal is to design and implement at an application that extracts these information from the fingerprint , which it graphically displays , statistically processes and provides them for further processing . Testing was performed using two synthetic databases , one real database and several separate fingerprints . The extracted information is compared with the gender of the fingerprint owner and it is evaluated whether it correlates together . Information showed some correlation , but due to the insufficiently large database , general conclusions cannot be made .
|
|
|
Detection of Papillary Line Width by Fingerprints
Homola, Antonín ; Drahanský, Martin (referee) ; Lodrová, Dana (advisor)
This work outlines a method of detection of the papillary line width in fingerprints. This method is one of the possible methods of liveness detection. The first part of the work with deals defining of the fingerprint, attacks on today's systems and possibilities to improve security. The next section detection describes of the papillary line width. During the process of resolving, the first thing to do was to start operation of the scanning device and to read the database for tests and experiments. An independent application was created on this purpose. Further, there were projected methods for detection and measuring of the papillary line width. Use of the Canny edge detector with the Sobel operator and the Gaussian filter proved the best. Then, there is described implementation of individual methods. The next part of the work describes and assesses the results of the tests. The last chapter summarizes the work and proposes further possibilities of development.
|
|
|
Synthetic Fingerprint Generation Using GAN
Dvořák, Jiří ; Drahanský, Martin (referee) ; Kanich, Ondřej (advisor)
Tato bakalářská práce se zabývá generováním syntetických otisků prstů za pomoci modelu založeném na principu generativních soupeřících sítí. Práce shrnuje základní teoretické informace z biometrie se zaměřením na otisky prstů. Zaobírá se také principem jednoho z populárních generátorů syntetických otisků prstů - nástrojem SFinGe. Práce představuje model postavený na hluboké konvoluční generativní soupeřící síti a představuje několik metod, které vedly ke zlepšení jeho výkonu. Vyhodnocení výsledků bylo provedeno výpočtem "Fréchet Inception Distance mezi vygenerovanými a existujími otisky. Dále byl vygenerován dataset obsahující 100 snímků. Ten byl vyhodnocen nástrojem NFIQ 2.0, který ukázal, že model je schopný generovat otisky prstů kvality srovnatelné s reálnými trénovacími daty.
|
|
|
Improvement of Methods for Detection and Classification of Damages in Fingerprint Images
Foltyn, Lukáš ; Heidari, Mona (referee) ; Kanich, Ondřej (advisor)
Cílem této práce je vylepšit stávající metody detekce a klasifikace poškození na snímcích otisků prstů s využitím předchozích prací studentů Vysokého učení technického v Brně. Práce je postavena na třech aplikacích: generátoru čárového poškození (jizvy, vlasy, rýhy), generátoru vlhkosti a aplikaci poskytující více různých modelů pro detekci a klasifikaci poškození otisků prstů. Pro vylepšení byly vybrány tři nejpřesnější modely - Faster-RCNN ResNet50, Faster-RCNN ResNet101 a CenterNet ResNet101. Práce popisuje vytvoření datové sady pomocí nepoškozených syntetických snímků otisků prstů s výše uvedenými uměle zavedenými poškozeními. Snaha o zlepšení přesnosti predikce modelů byla založena na přesnějším anotovaní ohraničujících boxů a úpravě hyperparametrů. Přestože práce přinesla určitá zlepšení, výsledky nejsou konzistentně úspěšné u všech modelů a typů poškození.
|
guest ::
