Handwriting features based detection of fake signatures: Short-Paper

Anton Akusok; Leonardo Espinosa-Leal; Kaj-Mikael Björk; Amaury Lendasse; Renjie Hu

doi:10.1145/3453892.3454003

Handwriting features based detection of fake signatures: Short-Paper

Anton Akusok, Leonardo Espinosa-Leal, Kaj-Mikael Björk, Amaury Lendasse, Renjie Hu

Not published at VTT

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

Abstract

Detection of fake signatures is a hard task. In this paper, we present a novel method for detecting trained forgeries using features extracted from sliding windows with different overlaps on a public available dataset of static images of signatures. Using a linear machine learning model named Extreme Learning Machine (ELM), our methodology achieves, in average, an Equal Error Rates (EER) of 2.31% for an overlap of 90%. In line with the state-of-the-art results available in the scientific literature.

Original language	English
Title of host publication	Proceedings of the 14th PErvasive Technologies Related to Assistive Environments Conference
Subtitle of host publication	PETRA '21
Editors	Filidia Makedon
Publisher	Association for Computing Machinery ACM
Pages	86-89
Number of pages	4
ISBN (Print)	978-1-4503-8792-7
DOIs	https://doi.org/10.1145/3453892.3454003
Publication status	Published - 29 Jun 2021
MoE publication type	A4 Article in a conference publication
Event	4th PErvasive Technologies Related to Assistive Environments Conference (PETRA '21) - Corfu, Greece Duration: 29 Jun 2021 → 2 Jul 2021

Conference

Conference	4th PErvasive Technologies Related to Assistive Environments Conference (PETRA '21)
Country/Territory	Greece
City	Corfu
Period	29/06/21 → 2/07/21

Keywords

Signature verification
biometrics
neural networks

Access to Document

10.1145/3453892.3454003

Cite this

@inproceedings{0cce14c2ddee4053b2c5782eb6730ae7,

title = "Handwriting features based detection of fake signatures: Short-Paper",

abstract = "Detection of fake signatures is a hard task. In this paper, we present a novel method for detecting trained forgeries using features extracted from sliding windows with different overlaps on a public available dataset of static images of signatures. Using a linear machine learning model named Extreme Learning Machine (ELM), our methodology achieves, in average, an Equal Error Rates (EER) of 2.31% for an overlap of 90%. In line with the state-of-the-art results available in the scientific literature.",

keywords = "Signature verification, biometrics, neural networks",

author = "Anton Akusok and Leonardo Espinosa-Leal and Kaj-Mikael Bj{\"o}rk and Amaury Lendasse and Renjie Hu",

year = "2021",

month = jun,

day = "29",

doi = "10.1145/3453892.3454003",

language = "English",

isbn = "978-1-4503-8792-7",

pages = "86--89",

editor = "Filidia Makedon",

booktitle = "Proceedings of the 14th PErvasive Technologies Related to Assistive Environments Conference",

publisher = "Association for Computing Machinery ACM",

address = "United States",

note = "4th PErvasive Technologies Related to Assistive Environments Conference (PETRA '21) ; Conference date: 29-06-2021 Through 02-07-2021",

}

Akusok, A, Espinosa-Leal, L, Björk, K-M, Lendasse, A & Hu, R 2021, Handwriting features based detection of fake signatures: Short-Paper. in F Makedon (ed.), Proceedings of the 14th PErvasive Technologies Related to Assistive Environments Conference: PETRA '21. Association for Computing Machinery ACM, pp. 86-89, 4th PErvasive Technologies Related to Assistive Environments Conference (PETRA '21), Corfu, Greece, 29/06/21. https://doi.org/10.1145/3453892.3454003

Handwriting features based detection of fake signatures: Short-Paper. / Akusok, Anton; Espinosa-Leal, Leonardo; Björk, Kaj-Mikael et al.
Proceedings of the 14th PErvasive Technologies Related to Assistive Environments Conference: PETRA '21. ed. / Filidia Makedon. Association for Computing Machinery ACM, 2021. p. 86-89.

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

TY - GEN

T1 - Handwriting features based detection of fake signatures

T2 - 4th PErvasive Technologies Related to Assistive Environments Conference (PETRA '21)

AU - Akusok, Anton

AU - Espinosa-Leal, Leonardo

AU - Björk, Kaj-Mikael

AU - Lendasse, Amaury

AU - Hu, Renjie

PY - 2021/6/29

Y1 - 2021/6/29

N2 - Detection of fake signatures is a hard task. In this paper, we present a novel method for detecting trained forgeries using features extracted from sliding windows with different overlaps on a public available dataset of static images of signatures. Using a linear machine learning model named Extreme Learning Machine (ELM), our methodology achieves, in average, an Equal Error Rates (EER) of 2.31% for an overlap of 90%. In line with the state-of-the-art results available in the scientific literature.

AB - Detection of fake signatures is a hard task. In this paper, we present a novel method for detecting trained forgeries using features extracted from sliding windows with different overlaps on a public available dataset of static images of signatures. Using a linear machine learning model named Extreme Learning Machine (ELM), our methodology achieves, in average, an Equal Error Rates (EER) of 2.31% for an overlap of 90%. In line with the state-of-the-art results available in the scientific literature.

KW - Signature verification

KW - biometrics

KW - neural networks

UR - http://www.scopus.com/inward/record.url?scp=85109343584&partnerID=8YFLogxK

U2 - 10.1145/3453892.3454003

DO - 10.1145/3453892.3454003

M3 - Conference article in proceedings

SN - 978-1-4503-8792-7

SP - 86

EP - 89

BT - Proceedings of the 14th PErvasive Technologies Related to Assistive Environments Conference

A2 - Makedon, Filidia

PB - Association for Computing Machinery ACM

Y2 - 29 June 2021 through 2 July 2021

ER -

Handwriting features based detection of fake signatures: Short-Paper

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this