An incremental learning method for spoof fingerprint detection

Jun Beom Kho, Wonjune Lee, Heeseung Choi, Jaihie Kim

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Spoof fingerprint detectors based on static features are built by learning a set of live and fake fingerprint images. These learning-based spoof detectors cannot accurately classify new or untrained types of fakes. To handle this problem, the existing spoof detector should be incrementally trained on the new types of fakes. This paper proposes a new spoof detection framework to learn new types of fakes incrementally without retraining the existing spoof detector repeatedly. The proposed model discriminates the newly learned fakes without serious loss of performance for the previously learned fakes and at the same time provides promising detection results for the various types of fakes. The proposed spoof detector integrates multiple “experts,” each of which shares the same structure but is separately trained for a different set of fake fingerprints. To detect a new type of fake fingerprint, a new expert exclusively trained on the new fake type is integrated into the spoof detector. Each expert consists of multiple support vector machines (SVMs) applied by an incremental learning algorithm (Learn++.NC), where each SVM adopts one of three texture features for spoof detection. Experimental results show the superiority of the proposed method compared with other methods in various scenarios.

Original languageEnglish
Pages (from-to)52-64
Number of pages13
JournalExpert Systems with Applications
Volume116
DOIs
Publication statusPublished - 2019 Feb 1

Fingerprint

Detectors
Support vector machines
Learning algorithms
Textures

All Science Journal Classification (ASJC) codes

  • Engineering(all)
  • Computer Science Applications
  • Artificial Intelligence

Cite this

Kho, Jun Beom ; Lee, Wonjune ; Choi, Heeseung ; Kim, Jaihie. / An incremental learning method for spoof fingerprint detection. In: Expert Systems with Applications. 2019 ; Vol. 116. pp. 52-64.
@article{39550cac0e6749beb7c65014c173c553,
title = "An incremental learning method for spoof fingerprint detection",
abstract = "Spoof fingerprint detectors based on static features are built by learning a set of live and fake fingerprint images. These learning-based spoof detectors cannot accurately classify new or untrained types of fakes. To handle this problem, the existing spoof detector should be incrementally trained on the new types of fakes. This paper proposes a new spoof detection framework to learn new types of fakes incrementally without retraining the existing spoof detector repeatedly. The proposed model discriminates the newly learned fakes without serious loss of performance for the previously learned fakes and at the same time provides promising detection results for the various types of fakes. The proposed spoof detector integrates multiple “experts,” each of which shares the same structure but is separately trained for a different set of fake fingerprints. To detect a new type of fake fingerprint, a new expert exclusively trained on the new fake type is integrated into the spoof detector. Each expert consists of multiple support vector machines (SVMs) applied by an incremental learning algorithm (Learn++.NC), where each SVM adopts one of three texture features for spoof detection. Experimental results show the superiority of the proposed method compared with other methods in various scenarios.",
author = "Kho, {Jun Beom} and Wonjune Lee and Heeseung Choi and Jaihie Kim",
year = "2019",
month = "2",
day = "1",
doi = "10.1016/j.eswa.2018.08.055",
language = "English",
volume = "116",
pages = "52--64",
journal = "Expert Systems with Applications",
issn = "0957-4174",
publisher = "Elsevier Limited",

}

An incremental learning method for spoof fingerprint detection. / Kho, Jun Beom; Lee, Wonjune; Choi, Heeseung; Kim, Jaihie.

In: Expert Systems with Applications, Vol. 116, 01.02.2019, p. 52-64.

Research output: Contribution to journalArticle

TY - JOUR

T1 - An incremental learning method for spoof fingerprint detection

AU - Kho, Jun Beom

AU - Lee, Wonjune

AU - Choi, Heeseung

AU - Kim, Jaihie

PY - 2019/2/1

Y1 - 2019/2/1

N2 - Spoof fingerprint detectors based on static features are built by learning a set of live and fake fingerprint images. These learning-based spoof detectors cannot accurately classify new or untrained types of fakes. To handle this problem, the existing spoof detector should be incrementally trained on the new types of fakes. This paper proposes a new spoof detection framework to learn new types of fakes incrementally without retraining the existing spoof detector repeatedly. The proposed model discriminates the newly learned fakes without serious loss of performance for the previously learned fakes and at the same time provides promising detection results for the various types of fakes. The proposed spoof detector integrates multiple “experts,” each of which shares the same structure but is separately trained for a different set of fake fingerprints. To detect a new type of fake fingerprint, a new expert exclusively trained on the new fake type is integrated into the spoof detector. Each expert consists of multiple support vector machines (SVMs) applied by an incremental learning algorithm (Learn++.NC), where each SVM adopts one of three texture features for spoof detection. Experimental results show the superiority of the proposed method compared with other methods in various scenarios.

AB - Spoof fingerprint detectors based on static features are built by learning a set of live and fake fingerprint images. These learning-based spoof detectors cannot accurately classify new or untrained types of fakes. To handle this problem, the existing spoof detector should be incrementally trained on the new types of fakes. This paper proposes a new spoof detection framework to learn new types of fakes incrementally without retraining the existing spoof detector repeatedly. The proposed model discriminates the newly learned fakes without serious loss of performance for the previously learned fakes and at the same time provides promising detection results for the various types of fakes. The proposed spoof detector integrates multiple “experts,” each of which shares the same structure but is separately trained for a different set of fake fingerprints. To detect a new type of fake fingerprint, a new expert exclusively trained on the new fake type is integrated into the spoof detector. Each expert consists of multiple support vector machines (SVMs) applied by an incremental learning algorithm (Learn++.NC), where each SVM adopts one of three texture features for spoof detection. Experimental results show the superiority of the proposed method compared with other methods in various scenarios.

UR - http://www.scopus.com/inward/record.url?scp=85053175036&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85053175036&partnerID=8YFLogxK

U2 - 10.1016/j.eswa.2018.08.055

DO - 10.1016/j.eswa.2018.08.055

M3 - Article

AN - SCOPUS:85053175036

VL - 116

SP - 52

EP - 64

JO - Expert Systems with Applications

JF - Expert Systems with Applications

SN - 0957-4174

ER -