Maximum-likelihood deformation analysis of different-sized fingerprints

  • Yuliang He*
  • , Jie Tian
  • , Qun Ren
  • , Xin Yang
  • *Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingChapterpeer-review

Abstract

This paper introduces a probabilistic formulation in terms of Maximum-likelihood estimation to calculate the optimal deformation parameters, such as scale, rotation and translation, between a pair of fingerprints acquired by different image capturers from the same finger. This uncertainty estimation technique allows parameter selection to be performed by choosing parameters that minimize the deformations uncertainty and maximize the global similarity between the pair of fingerprints. In addition, we use a multi-resolution search strategy to calculate the optimal deformation parameters in the space of possible deformation parameters. We apply the method to fingerprint matching in a pension fund management system in China, a fingerprint-based personal identification application system. The performance of the method shows that it is effective in estimating the optimal deformation parameters between a pair of fingerprints.

Original languageEnglish
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
EditorsJosef Kittler, Mark S. Nixon
PublisherSpringer Verlag
Pages421-428
Number of pages8
ISBN (Electronic)9783540403029
DOIs
StatePublished - 2003
Externally publishedYes

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume2688
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Fingerprint

Dive into the research topics of 'Maximum-likelihood deformation analysis of different-sized fingerprints'. Together they form a unique fingerprint.

Cite this