Publication Details

Psoriasis Damage Simulation into Synthetic Fingerprint

KANICH, O.; KOŠŤÁK, D.; DRAHANSKÝ, M. Psoriasis Damage Simulation into Synthetic Fingerprint. In Proceedings 18th International Conference of the Biometrics Special Interest Group. Lecture Notes in Informatics. Darmstadt: GI - Group for computer science, 2019. p. 205-212. ISBN: 978-3-88579-690-9.
Czech title
Simulace poškození lupénkou do syntetického otisku prstu
Type
conference paper
Language
English
Authors
Kanich Ondřej, Ing., Ph.D. (FIT)
Košťák David, Bc.
Drahanský Martin, prof. Ing., Ph.D.
Keywords

synthetic fingerprint, psoriasis, skin diseases, damage simulation, machine learning

Abstract

The goal of this article is to describemethod for simulation of damage done by psoriasis. Designed method is based onextracting subjects from real images and then including them into syntheticimages. Images are damaged by six different settings. Each setting representsdifferent level of disease severity. Results were verified by visual comparisonwith real images, consultation with medical doctor, quality measurement methods(NFIQ, VeriFinger, Oravec), and comparison score (VeriFinger). The most severedamage achieved median score of 38% (from the reference).

Published
2019
Pages
205–212
Proceedings
Proceedings 18th International Conference of the Biometrics Special Interest Group
Series
Lecture Notes in Informatics
Conference
18th International Conference of the Biometrics Special Interest Group, Darmstadt, DE
ISBN
978-3-88579-690-9
Publisher
GI - Group for computer science
Place
Darmstadt
UT WoS
000682778500006
EID Scopus
2-s2.0-85075890795
BibTeX 
@inproceedings{BUT159962,
  author="Ondřej {Kanich} and David {Košťák} and Martin {Drahanský}",
  title="Psoriasis Damage Simulation into Synthetic Fingerprint",
  booktitle="Proceedings 18th International Conference of the Biometrics Special Interest Group",
  year="2019",
  series="Lecture Notes in Informatics",
  pages="205--212",
  publisher="GI - Group for computer science",
  address="Darmstadt",
  isbn="978-3-88579-690-9",
  url="https://www.fit.vut.cz/research/publication/11258/"
}
Files
Back to top