Development of Pen-based ubiquitous interface for visually disabled person
| Project/Area Number |
15500401
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Rehabilitation science/Welfare engineering
|
|---|
| Research Institution | Kumamoto National College of Technology |
|---|
Principal Investigator |
KIYOTA Kimiyasu Kumamoto National College of Technology, associate Prof., 情報通信工学科, 助教授 (80186353)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
KOSHI Kazuhiro Kumamoto National College of Technology, associate Prof., 情報通信工学科, 助教授 (20303711)
EZAKI Nobuo Toba maritime College of Technology, Lecturer., 制御情報工学科, 講師 (30311038)
|
|---|
| Project Period (FY) |
2003 – 2004
|
| Project Status |
Completed (Fiscal Year 2004)
|
| Budget Amount *help |
¥3,700,000 (Direct Cost: ¥3,700,000)
Fiscal Year 2004: ¥1,400,000 (Direct Cost: ¥1,400,000)
Fiscal Year 2003: ¥2,300,000 (Direct Cost: ¥2,300,000)
|
|---|
| Keywords | Ubiquitous / Visually disabled person / PDA / Pen-based interface / Mobile PC / Character input system / ペン入力 / ヒューマンコミュニケーション / コミュニケーションエイド / 感覚代行 |
|---|
| Research Abstract |
We propose a pen-based NotePAD system on PDA (Personal Digital Assistance) as the ubiquitous information device for visually impaired person (particularly persons with acquired blindness). A blind user of this system is able to input Japanese characters directly instead of a keyboard without training. The system can recognize 3,126 characters including JIS Level 1 characters. This paper describes system design concept of our developed pen-based notepad system on PDA for blind person. We also investigate an improvement method of the recognition accuracy based on a Bi-gram probabilistic language model, and voice assistance function for blind person. The screen information is given to the user via built-in voice synthesizer in this system. From the comparison of our previous recognition method, the recognition accuracy of the 1st candidate has been improved to 84.7 % from 60.2 % in the proposed recognition system with applying the additional error correction method on PDA. Consequently, we confirmed the practical use possibility of the proposed system.
|
Report
(3 results)
Research Products
(19 results)
