DocumentCode
261635
Title
ANN based fingerprint image ROI segmentation
Author
Stojanovic, Branka ; Neskovic, Aleksandar ; Popovic, Zdravko ; Lukic, Vojislav
Author_Institution
Vlatacom d.o.o., Belgrade, Serbia
fYear
2014
fDate
25-27 Nov. 2014
Firstpage
505
Lastpage
508
Abstract
This paper proposes ANN based method for fingerprint ROI (Region of Interest) segmentation. Proposed ANNs where trained with 10000 samples extracted from 20 fingerprint images (in grey-scale and binary modes). The experimental results, including three statistical performance indicators, shows very good performance of the proposed method on a test database of 200 fingerprint images.
Keywords
feature extraction; fingerprint identification; image segmentation; neural nets; statistical analysis; ANN based fingerprint image ROI segmentation; ANN based method; binary modes; grey-scale modes; region of interest segmentation; statistical performance indicators; Artificial neural networks; Databases; Fingerprint recognition; Image matching; Image segmentation; Testing; Training; ANN; ROI segmentation; fingerprint;
fLanguage
English
Publisher
ieee
Conference_Titel
Telecommunications Forum Telfor (TELFOR), 2014 22nd
Conference_Location
Belgrade
Print_ISBN
978-1-4799-6190-0
Type
conf
DOI
10.1109/TELFOR.2014.7034457
Filename
7034457
Link To Document