|
|
Fingerprint biometry
Kubica, Roman ; Jiřík, Radovan (referee) ; Fedra, Petr (advisor)
The first part of this thesis deals with principles of evaluation biometric methods of scanning fingerprint images. Classification and description of single methods, separated into groups depending on the algorithm of comparison of template and image from database are included. In the next part is thesis focused on evaluation based on the success of experimental results and description of available databases. The main purpose of this thesis is creation of computer program for identification of fingerprint image in the programming interface MATLAB®. Realized identification algorithm is based on computing of Poincare index and extraction of singular points and minutiae. Performance of the program is tested on generated database and evaluated by index FAR and FRR.
|
|
|
Editor for WSQ Format of Fingerprint Images for Demonstration
Németh, Patrik ; Rydlo, Štěpán (referee) ; Kanich, Ondřej (advisor)
The goal of this thesis is to design and implement an application suitable for demonstrating the individual steps of fingerprint analysis, and which is able to automatically process fingerprints stored in the most widely used formats for storing fingerprint imagery, such as WSQ, PNG and JPEG 2000. In this thesis special attention was given to automatic processing methods and analysis of fingerprint imagery. It also offers an overview of the WSQ format and of its comparison to JPEG 2000. The output of this work is a graphical application capable of demonstrating individual steps of widely used methods of fingerprint imagery processing and analysis. The application implements Gabor filtering, line thinning based on the Zhang-Suen method, and other methods of fingerprint processing. With some error, it is capable of detecting minutiae, singularities and the class of a fingerprint. Some of the processing parameters may be modified by the user via the graphical interface. Processed images can be exported in various image formats.
|
|
|
Fingerprint Classification
Kostiha, Martin ; Orság, Filip (referee) ; Drahanský, Martin (advisor)
Automatic identification based on fingerprints requires the input fingerprint to be matched with a large number of fingerprints stored in a database. Fingerprint classification provides an important indexing mechanism in fingerprint databases that reduces the search time and computational complexity. This thesis is dealing with Galton-Henry's classification system. The classification method is based on orientation field and detection singularities using Poinaceré index.
|
|
|
Editor for WSQ Format of Fingerprint Images for Demonstration
Németh, Patrik ; Rydlo, Štěpán (referee) ; Kanich, Ondřej (advisor)
The goal of this thesis is to design and implement an application suitable for demonstrating the individual steps of fingerprint analysis, and which is able to automatically process fingerprints stored in the most widely used formats for storing fingerprint imagery, such as WSQ, PNG and JPEG 2000. In this thesis special attention was given to automatic processing methods and analysis of fingerprint imagery. It also offers an overview of the WSQ format and of its comparison to JPEG 2000. The output of this work is a graphical application capable of demonstrating individual steps of widely used methods of fingerprint imagery processing and analysis. The application implements Gabor filtering, line thinning based on the Zhang-Suen method, and other methods of fingerprint processing. With some error, it is capable of detecting minutiae, singularities and the class of a fingerprint. Some of the processing parameters may be modified by the user via the graphical interface. Processed images can be exported in various image formats.
|
|
|
Editor for WSQ Format of Fingerprint Images for Demonstration
Németh, Patrik ; Rydlo, Štěpán (referee) ; Kanich, Ondřej (advisor)
The goal of this thesis is to design and implement an application suitable for demonstrating the individual steps of fingerprint analysis, and which is able to automatically process fingerprints stored in the most widely used formats for storing fingerprint imagery, such as WSQ, PNG and JPEG 2000. In this thesis special attention was given to automatic processing methods and analysis of fingerprint imagery. It also offers an overview of the WSQ format and of its comparison to JPEG 2000. The output of this work is a graphical application capable of demonstrating individual steps of widely used methods of fingerprint imagery processing and analysis. The application implements Gabor filtering, line thinning based on the Zhang-Suen method, and other methods of fingerprint processing. With some error, it is capable of detecting minutiae, singularities and the class of a fingerprint. Some of the processing parameters may be modified by the user via the graphical interface. Processed images can be exported in various image formats.
|
|
|
Fingerprint biometry
Kubica, Roman ; Jiřík, Radovan (referee) ; Fedra, Petr (advisor)
The first part of this thesis deals with principles of evaluation biometric methods of scanning fingerprint images. Classification and description of single methods, separated into groups depending on the algorithm of comparison of template and image from database are included. In the next part is thesis focused on evaluation based on the success of experimental results and description of available databases. The main purpose of this thesis is creation of computer program for identification of fingerprint image in the programming interface MATLAB®. Realized identification algorithm is based on computing of Poincare index and extraction of singular points and minutiae. Performance of the program is tested on generated database and evaluated by index FAR and FRR.
|
|
|
Fingerprint Classification
Kostiha, Martin ; Orság, Filip (referee) ; Drahanský, Martin (advisor)
Automatic identification based on fingerprints requires the input fingerprint to be matched with a large number of fingerprints stored in a database. Fingerprint classification provides an important indexing mechanism in fingerprint databases that reduces the search time and computational complexity. This thesis is dealing with Galton-Henry's classification system. The classification method is based on orientation field and detection singularities using Poinaceré index.
|
guest ::
