| Project/Area Number |
03402036
|
|---|
| Research Category |
Grant-in-Aid for General Scientific Research (A)
|
| Allocation Type | Single-year Grants |
|---|
| Research Field |
情報工学
|
|---|
| Research Institution | Kyoto University |
|---|
Principal Investigator |
IKEDA Katsuo Kyoto Univ., Information Science, Prof., 工学部, 教授 (30026009)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
OHTA Yuichi Univ. of Tsukuba, Information Sciences and Electronics, Prof., 電子・情報工学系, 教授 (50115804)
WATANABE Masako Kyoto Univ. Integ. Media Env. Exp. Lab, Res., 工学部, 教務職員 (70127158)
AMANO Akira Kyoto Univ., Information Science, Inst., 工学部, 助手 (60252491)
HIROSE Shouichi Kyoto Univ., Information Science, Inst., 工学部, 助手 (20228836)
MINOH Michihiko Kyoto Univ. Integ. Media Env. Exp. Lab, Assoc. Prof., 工学部, 助教授 (70166099)
|
|---|
| Project Period (FY) |
1991 – 1993
|
| Project Status |
Completed (Fiscal Year 1993)
|
| Budget Amount *help |
¥19,200,000 (Direct Cost: ¥19,200,000)
Fiscal Year 1993: ¥1,800,000 (Direct Cost: ¥1,800,000)
Fiscal Year 1992: ¥10,000,000 (Direct Cost: ¥10,000,000)
Fiscal Year 1991: ¥7,400,000 (Direct Cost: ¥7,400,000)
|
|---|
| Keywords | Symbolic Image Recognition / Constructive Inference / Multi-Dimensional Paral lel Processing |
|---|
| Research Abstract |
The aim of this research is to establish multi-dimensional parallel processing method for various symbolic image understanding and recognition, by constructing general constructive inference system with proper information of target symbolic image. We worked on the construction of the general constructive inference system, symbolic image specific vectorization, and automatic generation of inference rule. Also, we organaized the whole investigation. Results of the research is as follows :
1. In the symbolic image processing system, we need to handle various data such as bitmap images and graphs. In order to handle them efficiently, we made a library which dynamically constructs and rearranges the arrays, lists and structured data.
2. We developed the candidates verification system with ATMS method. The system generates candidates of vectors, paraller lines or loops with image operation, them by their locational relations.
3. We proposed a method of recognition of line drawn face. In the system, parts of a face eyes, ears are extracted in parallel, then verifies them by their locational relations.
4. We developed a system which generates inferencing rules by inductive learning from sample symbolic images. By generating these rules, we can use general symbolic image processing system.
5. We developed a image processing system which extracts regions or lines satisfying requests from image recognition system.
|
