Unsupervised Fingerprint Recognition

Wei-Ho TSAI; Jun-Wei LIN; Der-Chang TSENG

doi:10.1587/transinf.E96.D.2115

Unsupervised Fingerprint Recognition

Wei-Ho TSAI, Jun-Wei LIN, Der-Chang TSENG

Full Text Views

0

Cite this

Summary :

This study extends conventional fingerprint recognition from a supervised to an unsupervised framework. Instead of enrolling fingerprints from known persons to identify unknown fingerprints, our aim is to partition a collection of unknown fingerprints into clusters, so that each cluster consists of fingerprints from the same finger and the number of generated clusters equals the number of distinct fingers involved in the collection. Such an unsupervised framework is helpful to handle the situation where a collection of captured fingerprints are not from the enrolled people. The task of fingerprint clustering is formulated as a problem of minimizing the clustering errors characterized by the Rand index. We estimate the Rand index by computing the similarities between fingerprints and then apply a genetic algorithm to minimize the Rand index. Experiments conducted using the FVC2002 database show that the proposed fingerprint clustering method outperforms an intuitive method based on hierarchical agglomerative clustering. The experiments also show that the number of clusters determined by our system is close to the true number of distinct fingers involved in the collection.

Publication: IEICE TRANSACTIONS on Information Vol.E96-D No.9 pp.2115-2125

Publication Date: 2013/09/01

Publicized

Online ISSN: 1745-1361

DOI: 10.1587/transinf.E96.D.2115

Type of Manuscript: PAPER

Category: Image Recognition, Computer Vision

Authors

Wei-Ho TSAI
  National Taipei University of Technology
Jun-Wei LIN
  National Taipei University of Technology
Der-Chang TSENG
  National Taipei University of Technology

Keyword

clustering, fingerprint, Rand index, unsupervised

Cite this

Copy

Wei-Ho TSAI, Jun-Wei LIN, Der-Chang TSENG, "Unsupervised Fingerprint Recognition" in IEICE TRANSACTIONS on Information, vol. E96-D, no. 9, pp. 2115-2125, September 2013, doi: 10.1587/transinf.E96.D.2115.
Abstract: This study extends conventional fingerprint recognition from a supervised to an unsupervised framework. Instead of enrolling fingerprints from known persons to identify unknown fingerprints, our aim is to partition a collection of unknown fingerprints into clusters, so that each cluster consists of fingerprints from the same finger and the number of generated clusters equals the number of distinct fingers involved in the collection. Such an unsupervised framework is helpful to handle the situation where a collection of captured fingerprints are not from the enrolled people. The task of fingerprint clustering is formulated as a problem of minimizing the clustering errors characterized by the Rand index. We estimate the Rand index by computing the similarities between fingerprints and then apply a genetic algorithm to minimize the Rand index. Experiments conducted using the FVC2002 database show that the proposed fingerprint clustering method outperforms an intuitive method based on hierarchical agglomerative clustering. The experiments also show that the number of clusters determined by our system is close to the true number of distinct fingers involved in the collection.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.E96.D.2115/_p

Copy

@ARTICLE{e96-d_9_2115,
author={Wei-Ho TSAI, Jun-Wei LIN, Der-Chang TSENG, },
journal={IEICE TRANSACTIONS on Information},
title={Unsupervised Fingerprint Recognition},
year={2013},
volume={E96-D},
number={9},
pages={2115-2125},
abstract={This study extends conventional fingerprint recognition from a supervised to an unsupervised framework. Instead of enrolling fingerprints from known persons to identify unknown fingerprints, our aim is to partition a collection of unknown fingerprints into clusters, so that each cluster consists of fingerprints from the same finger and the number of generated clusters equals the number of distinct fingers involved in the collection. Such an unsupervised framework is helpful to handle the situation where a collection of captured fingerprints are not from the enrolled people. The task of fingerprint clustering is formulated as a problem of minimizing the clustering errors characterized by the Rand index. We estimate the Rand index by computing the similarities between fingerprints and then apply a genetic algorithm to minimize the Rand index. Experiments conducted using the FVC2002 database show that the proposed fingerprint clustering method outperforms an intuitive method based on hierarchical agglomerative clustering. The experiments also show that the number of clusters determined by our system is close to the true number of distinct fingers involved in the collection.},
keywords={},
doi={10.1587/transinf.E96.D.2115},
ISSN={1745-1361},
month={September},}

Copy

TY - JOUR
TI - Unsupervised Fingerprint Recognition
T2 - IEICE TRANSACTIONS on Information
SP - 2115
EP - 2125
AU - Wei-Ho TSAI
AU - Jun-Wei LIN
AU - Der-Chang TSENG
PY - 2013
DO - 10.1587/transinf.E96.D.2115
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E96-D
IS - 9
JA - IEICE TRANSACTIONS on Information
Y1 - September 2013
AB - This study extends conventional fingerprint recognition from a supervised to an unsupervised framework. Instead of enrolling fingerprints from known persons to identify unknown fingerprints, our aim is to partition a collection of unknown fingerprints into clusters, so that each cluster consists of fingerprints from the same finger and the number of generated clusters equals the number of distinct fingers involved in the collection. Such an unsupervised framework is helpful to handle the situation where a collection of captured fingerprints are not from the enrolled people. The task of fingerprint clustering is formulated as a problem of minimizing the clustering errors characterized by the Rand index. We estimate the Rand index by computing the similarities between fingerprints and then apply a genetic algorithm to minimize the Rand index. Experiments conducted using the FVC2002 database show that the proposed fingerprint clustering method outperforms an intuitive method based on hierarchical agglomerative clustering. The experiments also show that the number of clusters determined by our system is close to the true number of distinct fingers involved in the collection.
ER -