| Project/Area Number |
17300267
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (B)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Educational technology
|
|---|
| Research Institution | Mie University |
|---|
Principal Investigator |
TSURUOKA Shinji Mie University, Electrical and Electronic Engineering, Professor (30126982)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
SHINOGI Tsuyoshi Mie University, Electrical and Electronic Engineering, Professor (90303750)
YOSHIKAWA Tomohiro Nagoya University, Computational Science and Engineering, Associate Professor (20303753)
KAWANAKA Hiroharu Mie University, Electrical and Electronic Engineering, Assistant Professor (30437115)
|
|---|
| Project Period (FY) |
2005 – 2007
|
| Project Status |
Completed (Fiscal Year 2007)
|
| Budget Amount *help |
¥12,770,000 (Direct Cost: ¥12,500,000、Indirect Cost: ¥270,000)
Fiscal Year 2007: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
Fiscal Year 2006: ¥3,000,000 (Direct Cost: ¥3,000,000)
Fiscal Year 2005: ¥8,600,000 (Direct Cost: ¥8,600,000)
|
|---|
| Keywords | e-learning / document recognition / behavior recognition / document database retrieval / electronic document / handwritten recognition / コンピュータ支援教育 / 教科書データベース / 手書き文字認識 / キーワード検索 / 高速化アルゴリズム / 知識メディア / 視野決定システム / 単語辞書 / ソフトウェアの高速化 / カメラ制御 |
|---|
| Research Abstract |
This research project developed the following challenging approaches.
(1) Dynamic determination method in the view of an active camera in large rlacc and scene segmentation for e-leaning
Anew algorithm based on the behavior of a teacher is proposed for a dynamic determination method in the view of an active camera in large class, and the algorithm reduced the motion time of camera view. Two approaches are proposed for scene segmentation for generation the interactive e-learning contents. The first approach is based on the blackboard state, and the second approach is based on anon-voice sound.
(2) Handwritten keyword recognition on the blackboard and the generation of high resolution image
Handwritten keyword recognition algorithms are developed using keyword lexicon for the characters on the blackboard. A new kcal threshold method "median differential method" and a new generating method "genetic image merge algorithm" of the high resolution image are proposed for the blackboard image in large class
(3) Digitizing method for paper documents and the development of an individual interactive education system
In higher education, the student interests in the various contents in real world, but a lot of required information is not opened for the digitizing coat The lack of the information is one of important problems in e-learning. We are challenging the automatic translation from paper documents to digital documents, and we developed. a new extracting the document structure from a paper document. And we developed the individual interactive education system licitly the translated digital documents
|
