Development of expert system "NYMPH" for identification of trees and its realization
| Project/Area Number |
04660160
|
|---|
| Research Category |
Grant-in-Aid for General Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Research Field |
林学
|
|---|
| Research Institution | Gifu University |
|---|
Principal Investigator |
OJIKA Takeo Gifu Univ., Fac.Eng., Prof., 工学部, 教授 (80030451)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
IZUMOTO Yoshoro Osaka Kyoiku Univ., Fac.Educ., Assoc.Prof., 教育学部, 助教授 (70030393)
HARAYAMA Michiko Gifu Univ., Fac.Eng., Assist.Prof., 工学部, 助手 (30237614)
KOMIYAMA Akira Gifu Univ., Fac.Agr., Assoc.Prof., 農学部, 助教授 (60135184)
SHINODA Yoshihiko Gifu Univ., Grad.Schl.Agr.Sci., Prof., 大学院・連合農学研究科, 教授 (50021712)
|
|---|
| Project Period (FY) |
1992 – 1993
|
| Project Status |
Completed (Fiscal Year 1993)
|
| Budget Amount *help |
¥2,100,000 (Direct Cost: ¥2,100,000)
Fiscal Year 1993: ¥1,100,000 (Direct Cost: ¥1,100,000)
Fiscal Year 1992: ¥1,000,000 (Direct Cost: ¥1,000,000)
|
|---|
| Keywords | Identification of trees / Expert system / Leaf / Production rule / Image processing / C-language / Divided leaf / Outline shape / 樹木 / 識別 / 円形度 / k曲率 / 自動認識 / 樹種識別 / エキスパート・シェル / ファジィ推論 / 手続き型言語 |
|---|
| Research Abstract |
The aim of the present research is to develope an expert system "NYMPH" for identification of trees from a leaf. It is first necessary to develop an efficient expert shell. For that we developed SC language which has an affinity for C programs. The SC shell is an extended language which allows the object-oriented programming and the inference based on the production-rule know ledge base. The objects are defined by the class statement, and the inference is triggered as well as when functions are referred in the program.
Then, by using the SC shell, we established the knowledge base as for leaf-characteristics by automatic image processing such as no. of divided leaves, lenghth, shapes, pattern of veins, and so on. According to the present expert system "NYMPH", it is possible now to identify more than one hundred kinds of familiar trees from just a single leaf.
|
Report
(3 results)
Research Products
(4 results)
