Writer Identification for Offline Handwritten Kanji without using Character Recognition Features

被引：0

作者：

Soma, Ayumu ^{[1
]}

Arai, Masayuki ^{[1
]}

机构：

[1] Teikyo Univ, Grad Sch Sci & Engn, Utsunomiya, Tochigi, Japan

来源：

PROCEEDINGS OF THE 2013 INTERNATIONAL CONFERENCE ON INFORMATION SCIENCE AND TECHNOLOGY APPLICATIONS (ICISTA-2013) | 2013年 / 58卷

关键词：

writer identification; handwritten characters; offline characters; Kanji; VERIFICATION;

D O I：

暂无

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Most research on writer identification in the case of offline handwritten Kanji characters uses character recognition features. In this paper, we assume the following is representative of the writers' style: the start point, the end point, the angle of each stroke composing a Kanji character, and the size and position of the Kanji character. The experimental results show that the identification rates are 95.2% without rejection and 99.6% with 10% rejection for four Kanji characters written by one hundred writers.

引用

页码：96 / 98

页数：3

共 6 条

[1] Text-independent writer identification and verification using textural and allographic features [J].

Bulacu, Marius ;

Schomaker, Lambert .

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2007, 29 (04) :701-717

[2] Offline signature verification and identification using distance statistics [J].

Kalera, MK ;

Srihari, S ;

Xu, AH .

INTERNATIONAL JOURNAL OF PATTERN RECOGNITION AND ARTIFICIAL INTELLIGENCE, 2004, 18 (07) :1339-1360

[3]

Misaki K., 1997, T JAPANESE ASS FOREN, V2, P71

[4]

Misaki K., 2002, T JAPANESE ASS FOREN, V7, P71

[5]

Tokiwa K., 2011, T I IMAGE ELECT ENG, V40, P660

[6]

Umeda M., 2002, Transactions of the Institute of Electrical Engineers of Japan, Part C, V122-C, P1869

← 1 →