A new and effective method in fingerprint classification
Life Science Journal, cilt.10, sa.SPL.ISSUE 12, ss.584-588, 2013 (Scopus)
- Yayın Türü: Makale / Tam Makale
- Cilt numarası: 10 Sayı: SPL.ISSUE 12
- Basım Tarihi: 2013
- Dergi Adı: Life Science Journal
- Derginin Tarandığı İndeksler: Scopus
- Sayfa Sayıları: ss.584-588
- Anahtar Kelimeler: Classification, Co-occurrence matrix, Fingerprint
- Gazi Üniversitesi Adresli: Evet
Özet
Nowadays using fingerprint as a biometric feature is widely utilized in related applications of individuals' identification. Reliability, cost-effectiveness and it applicability in comparison to the other biometric IDs including signature, iris identification, face identification and motion identification, has led to the widespread use of fingerprint in different applications. If too much data is in hand fingerprint classification is one of the ways to reduce the time of searching and identifying an unknown image in a great set of images. In this paper a new method for classifying fingerprints in databases FVC 2000, FVC 2002 and FVC 2004 is presented. In order to extract features from fingerprint images we used co-occurrence matrix and in order to classifying and identifying individuals we used a neural network. The used neural network was a multilayer perceptron network with training after error propagation method.