Kernel deep regression network for touch-stroke dynamics authentication

Inho Chang, Cheng Yaw Low, Seokmin Choi, Beng Jin Teoh

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

Touch-stroke dynamics is an emerging behavioral biometrics justified feasible for mobile identity management. A touch-stroke dynamics authentication system is composed of a hand-engineered feature extractor and a classifier separately. In this letter, we propose a stacking-based deep learning network that performs feature extraction and classification, collectively dubbed Kernel Deep Regression Network (KDRN). The KDRN is built on multiple kernel ridge regressions (KRR) hierarchically, where each is trained analytically and independently. In principal, KDRN does not mean to learn directly from the raw touch-stroke data like other deep learning models, but it relearns from the pre-extracted features to yield a richer and a relatively more discriminative feature set. Subsequent to that, the authentication is carried out by KRR. Overall, KDRN achieves an equal error rate of 0.013% for intrasession authentication, 0.023% for intersession authentication, and 0.121% for interweek authentication on the Touchlaytics dataset.

Original languageEnglish
Pages (from-to)1109-1113
Number of pages5
JournalIEEE Signal Processing Letters
Volume25
Issue number7
DOIs
Publication statusPublished - 2018 Jul 1

Fingerprint

Kernel Regression
Stroke
Authentication
Ridge Regression
Identity Management
Biometrics
Extractor
Stacking
Feature extraction
Classifiers
Feature Extraction
Error Rate
Classifier

All Science Journal Classification (ASJC) codes

  • Signal Processing
  • Electrical and Electronic Engineering
  • Applied Mathematics

Cite this

Chang, Inho ; Low, Cheng Yaw ; Choi, Seokmin ; Teoh, Beng Jin. / Kernel deep regression network for touch-stroke dynamics authentication. In: IEEE Signal Processing Letters. 2018 ; Vol. 25, No. 7. pp. 1109-1113.
@article{6f20af3d554c4b5b94f5080ec8746e6a,
title = "Kernel deep regression network for touch-stroke dynamics authentication",
abstract = "Touch-stroke dynamics is an emerging behavioral biometrics justified feasible for mobile identity management. A touch-stroke dynamics authentication system is composed of a hand-engineered feature extractor and a classifier separately. In this letter, we propose a stacking-based deep learning network that performs feature extraction and classification, collectively dubbed Kernel Deep Regression Network (KDRN). The KDRN is built on multiple kernel ridge regressions (KRR) hierarchically, where each is trained analytically and independently. In principal, KDRN does not mean to learn directly from the raw touch-stroke data like other deep learning models, but it relearns from the pre-extracted features to yield a richer and a relatively more discriminative feature set. Subsequent to that, the authentication is carried out by KRR. Overall, KDRN achieves an equal error rate of 0.013{\%} for intrasession authentication, 0.023{\%} for intersession authentication, and 0.121{\%} for interweek authentication on the Touchlaytics dataset.",
author = "Inho Chang and Low, {Cheng Yaw} and Seokmin Choi and Teoh, {Beng Jin}",
year = "2018",
month = "7",
day = "1",
doi = "10.1109/LSP.2018.2846050",
language = "English",
volume = "25",
pages = "1109--1113",
journal = "IEEE Signal Processing Letters",
issn = "1070-9908",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "7",

}

Kernel deep regression network for touch-stroke dynamics authentication. / Chang, Inho; Low, Cheng Yaw; Choi, Seokmin; Teoh, Beng Jin.

In: IEEE Signal Processing Letters, Vol. 25, No. 7, 01.07.2018, p. 1109-1113.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Kernel deep regression network for touch-stroke dynamics authentication

AU - Chang, Inho

AU - Low, Cheng Yaw

AU - Choi, Seokmin

AU - Teoh, Beng Jin

PY - 2018/7/1

Y1 - 2018/7/1

N2 - Touch-stroke dynamics is an emerging behavioral biometrics justified feasible for mobile identity management. A touch-stroke dynamics authentication system is composed of a hand-engineered feature extractor and a classifier separately. In this letter, we propose a stacking-based deep learning network that performs feature extraction and classification, collectively dubbed Kernel Deep Regression Network (KDRN). The KDRN is built on multiple kernel ridge regressions (KRR) hierarchically, where each is trained analytically and independently. In principal, KDRN does not mean to learn directly from the raw touch-stroke data like other deep learning models, but it relearns from the pre-extracted features to yield a richer and a relatively more discriminative feature set. Subsequent to that, the authentication is carried out by KRR. Overall, KDRN achieves an equal error rate of 0.013% for intrasession authentication, 0.023% for intersession authentication, and 0.121% for interweek authentication on the Touchlaytics dataset.

AB - Touch-stroke dynamics is an emerging behavioral biometrics justified feasible for mobile identity management. A touch-stroke dynamics authentication system is composed of a hand-engineered feature extractor and a classifier separately. In this letter, we propose a stacking-based deep learning network that performs feature extraction and classification, collectively dubbed Kernel Deep Regression Network (KDRN). The KDRN is built on multiple kernel ridge regressions (KRR) hierarchically, where each is trained analytically and independently. In principal, KDRN does not mean to learn directly from the raw touch-stroke data like other deep learning models, but it relearns from the pre-extracted features to yield a richer and a relatively more discriminative feature set. Subsequent to that, the authentication is carried out by KRR. Overall, KDRN achieves an equal error rate of 0.013% for intrasession authentication, 0.023% for intersession authentication, and 0.121% for interweek authentication on the Touchlaytics dataset.

UR - http://www.scopus.com/inward/record.url?scp=85048504110&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85048504110&partnerID=8YFLogxK

U2 - 10.1109/LSP.2018.2846050

DO - 10.1109/LSP.2018.2846050

M3 - Article

AN - SCOPUS:85048504110

VL - 25

SP - 1109

EP - 1113

JO - IEEE Signal Processing Letters

JF - IEEE Signal Processing Letters

SN - 1070-9908

IS - 7

ER -