Contactless Fingerprint Recognition System Based On CNN

  • IshaMeshram ,P. P. Rathod, Manali Nangre, Pranali Kolhale

Abstract

Fingerprint recognition is a popular problem in the field of pattern recognition. It is majorly used in modern authentication technology like in access devices in mobile phones. The goal of this project is to investigate the applicability of convolutional neural networks for fingerprint recognition. This paper builds a CNN-based framework to precisely match contactless and contact-based fingerprint images. This framework initially trains a multi-Siamese CNN using fingerprint details, respective ridge map and specific area of ridge map. A distance-aware loss function is generated using deep fingerprint representation. For more accurate cross comparison deep fingerprint representations generated in such multi-Siamese network are concatenated. The proposed methodology for cross-fingerprint comparison is calculated on two publicly available data. The available database contains contactless 2D fingerprints and respective contact-based fingerprints.    

Published
2020-07-01