Off-line signature verification without requiring random forgeries for training

被引：0

作者：

Murshed, NA ^{[1
]}

Bortolozzi, F ^{[1
]}

Sabourin, R ^{[1
]}

机构：

[1] ECOLE TECHNOL SUPER,MONTREAL,PQ H2T 2C8,CANADA

来源：

IMAGE ANALYSIS APPLICATIONS AND COMPUTER GRAPHICS | 1995年 / 1024卷

关键词：

D O I：

暂无

中图分类号：

TP31 [计算机软件];

学科分类号：

081202 ; 0835 ;

摘要：

This paper presents an Off-line Signature Verification System for the elimination of random forgeries. Compared with the proposed systems thus far, our system is trained with genuine signatures only. This eliminates many of the problems existent in the current systems. The proposed system is evaluated with a data base of 200 signatures.

引用

页码：107 / 115

页数：9

共 7 条

[1] CARDOT H, 1992, ARTIFICIAL NEURAL NE
[2] COTTRELL GW, 1989, ADV COGNITIVE SCI, V3
[3] MCCORMACK DKR, 1992, NEURAL COMPUTING R 2, P243
[4] MIGHELL DA, 1989, ADV NEURAL INFORMATI, V1, P341
[5] Ostu Nobuyuki, 1979, IEEE T SYST MAN CYB, VSMC-9, P62, DOI DOI 10.1109/TSMC.1979.4310076
[6] AUTOMATIC SIGNATURE VERIFICATION AND WRITER IDENTIFICATION - THE STATE OF THE ART
PLAMONDON, R
LORETTE, G
[J]. PATTERN RECOGNITION, 1989, 22 (02) : 107 - 131
[7] Sabourin R., 1993, Proceedings of the Second International Conference on Document Analysis and Recognition (Cat. No.93TH0578-5), P1, DOI 10.1109/ICDAR.1993.395795

← 1 →