Japanese Historical Character Recognition using Hierarchical Classifiers and Its Applications
Project/Area Number |
20560389
|
Research Category |
Grant-in-Aid for Scientific Research (C)
|
Allocation Type | Single-year Grants |
Section | 一般 |
Research Field |
System engineering
|
Research Institution | Matsue National College of Technology |
Principal Investigator |
HORIUCHI Tadashi Matsue National College of Technology, 電子制御工学科, 准教授 (50294129)
|
Co-Investigator(Kenkyū-buntansha) |
KATO Satoru 松江工業高等専門学校, 情報工学科, 講師 (40342547)
|
Co-Investigator(Renkei-kenkyūsha) |
YAMAZAKI Masakatsu 比治山大学, 現代文化学部, 准教授 (10342544)
|
Project Period (FY) |
2008 – 2010
|
Project Status |
Completed (Fiscal Year 2010)
|
Budget Amount *help |
¥4,550,000 (Direct Cost: ¥3,500,000、Indirect Cost: ¥1,050,000)
Fiscal Year 2010: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
Fiscal Year 2009: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000)
Fiscal Year 2008: ¥2,210,000 (Direct Cost: ¥1,700,000、Indirect Cost: ¥510,000)
|
Keywords | パターン認識 / 機械学習 / 古文書文字認識 / 古文書読解支援 / サポートベクターマシン |
Research Abstract |
In this research, we developed the Japanese historical character recognition system using the hierarchical classifiers. The hierarchical classifiers consist of a rough-classifier and a set of fine-classifiers. Moreover, we developed the reading support system for Japanese historical documents. In this reading support system, the user selects the unknown character by mouse operation and the character recognition system outputs five candidate characters.
|
Report
(4 results)
Research Products
(25 results)