Multimodal biometric score fusion: The Mean Rule vs. Support Vector classifiers

Sonia Garcia-Salicetti; Mohamed Anouar Mellakh; Lorène Allano; Bernadette Dorizzi

Multimodal biometric score fusion: The Mean Rule vs. Support Vector classifiers

Sonia Garcia-Salicetti, Mohamed Anouar Mellakh, Lorène Allano, Bernadette Dorizzi

CNRS SAMOVAR UMR 5157

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Recently, a discrepancy in results has appeared in the literature concerning score fusion methods, classified in .combination methods. and .classification methods. [1]. Some works suggest that a simple Arithmetic Mean Rule (AMR) can outperform some training-based methods on multimodal data [2], while others favour, among other trained classifiers, a Support Vector Machine [3]. This paper makes a comparative study of the Arithmetic Mean Rule (AMR) coupled with different state-of-the-art normalization techniques [4, 5] and a linear Support Vector Machine (SVM), in the framework of voice and on-line signature scores fusion. Two experiments differing in the difficulty to discriminate genuine from impostor accesses are carried out on the BIOMET database [6].

Original language	English
Title of host publication	13th European Signal Processing Conference, EUSIPCO 2005
Pages	2282-2285
Number of pages	4
Publication status	Published - 1 Dec 2005
Externally published	Yes
Event	13th European Signal Processing Conference, EUSIPCO 2005 - Antalya, Turkey Duration: 4 Sept 2005 → 8 Sept 2005

Publication series

Name	13th European Signal Processing Conference, EUSIPCO 2005

Conference

Conference	13th European Signal Processing Conference, EUSIPCO 2005
Country/Territory	Turkey
City	Antalya
Period	4/09/05 → 8/09/05

Cite this

@inproceedings{8e5cde0eb2c944ad916bef3ed7cf8844,

title = "Multimodal biometric score fusion: The Mean Rule vs. Support Vector classifiers",

abstract = "Recently, a discrepancy in results has appeared in the literature concerning score fusion methods, classified in .combination methods. and .classification methods. [1]. Some works suggest that a simple Arithmetic Mean Rule (AMR) can outperform some training-based methods on multimodal data [2], while others favour, among other trained classifiers, a Support Vector Machine [3]. This paper makes a comparative study of the Arithmetic Mean Rule (AMR) coupled with different state-of-the-art normalization techniques [4, 5] and a linear Support Vector Machine (SVM), in the framework of voice and on-line signature scores fusion. Two experiments differing in the difficulty to discriminate genuine from impostor accesses are carried out on the BIOMET database [6].",

author = "Sonia Garcia-Salicetti and Mellakh, \{Mohamed Anouar\} and Lor{\`e}ne Allano and Bernadette Dorizzi",

year = "2005",

month = dec,

day = "1",

language = "English",

isbn = "1604238216",

series = "13th European Signal Processing Conference, EUSIPCO 2005",

pages = "2282--2285",

booktitle = "13th European Signal Processing Conference, EUSIPCO 2005",

note = "13th European Signal Processing Conference, EUSIPCO 2005 ; Conference date: 04-09-2005 Through 08-09-2005",

}

Multimodal biometric score fusion: The Mean Rule vs. Support Vector classifiers. / Garcia-Salicetti, Sonia; Mellakh, Mohamed Anouar; Allano, Lorène et al.
13th European Signal Processing Conference, EUSIPCO 2005. 2005. p. 2282-2285 (13th European Signal Processing Conference, EUSIPCO 2005).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Multimodal biometric score fusion

T2 - 13th European Signal Processing Conference, EUSIPCO 2005

AU - Garcia-Salicetti, Sonia

AU - Mellakh, Mohamed Anouar

AU - Allano, Lorène

AU - Dorizzi, Bernadette

PY - 2005/12/1

Y1 - 2005/12/1

N2 - Recently, a discrepancy in results has appeared in the literature concerning score fusion methods, classified in .combination methods. and .classification methods. [1]. Some works suggest that a simple Arithmetic Mean Rule (AMR) can outperform some training-based methods on multimodal data [2], while others favour, among other trained classifiers, a Support Vector Machine [3]. This paper makes a comparative study of the Arithmetic Mean Rule (AMR) coupled with different state-of-the-art normalization techniques [4, 5] and a linear Support Vector Machine (SVM), in the framework of voice and on-line signature scores fusion. Two experiments differing in the difficulty to discriminate genuine from impostor accesses are carried out on the BIOMET database [6].

AB - Recently, a discrepancy in results has appeared in the literature concerning score fusion methods, classified in .combination methods. and .classification methods. [1]. Some works suggest that a simple Arithmetic Mean Rule (AMR) can outperform some training-based methods on multimodal data [2], while others favour, among other trained classifiers, a Support Vector Machine [3]. This paper makes a comparative study of the Arithmetic Mean Rule (AMR) coupled with different state-of-the-art normalization techniques [4, 5] and a linear Support Vector Machine (SVM), in the framework of voice and on-line signature scores fusion. Two experiments differing in the difficulty to discriminate genuine from impostor accesses are carried out on the BIOMET database [6].

M3 - Conference contribution

AN - SCOPUS:84863707148

SN - 1604238216

SN - 9781604238211

T3 - 13th European Signal Processing Conference, EUSIPCO 2005

SP - 2282

EP - 2285

BT - 13th European Signal Processing Conference, EUSIPCO 2005

Y2 - 4 September 2005 through 8 September 2005

ER -

Multimodal biometric score fusion: The Mean Rule vs. Support Vector classifiers

Abstract

Publication series

Conference

Fingerprint

Cite this