2017 Fiscal Year Final Research Report
Non-destructive wood identification and evaluation of cultural properties and their archives.
| Project/Area Number |
25252033
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (A)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Wood science
|
|---|
| Research Institution | Kyoto University |
|---|
Principal Investigator |
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
今津 節生 奈良大学, 文学部, 教授 (50250379)
|
|---|
| Research Collaborator |
KOBAYASHI Kayoko
HWANG Sung Wook
KEGASA Takahiro
KANAI Izumi
TAZURU Suyako
SORIMACHI Hajime
TORIGOE Toshiyuki
AKADA Masatoshi
|
|---|
| Project Period (FY) |
2013-04-01 – 2018-03-31
|
| Keywords | 樹種識別 / 機械学習 / デジタル画像 / グレーレベル同時生起行列 / スケール不変特徴変換 / ディープラーニング / 光学顕微鏡 / X線CT |
|---|
| Outline of Final Research Achievements |
By replacing observer's eyes to artificial intelligence, we have made it possible to identify anatomically closely related species that were impossible so far on the basis of microscopic observation. At the same time, it enabled interdisciplinary development of identification technology to other fields. In addition to combinations of feature extraction and discriminators designed by humans, deep learning that leaves all processes from feature extraction to intellectual judgment to multilayered neural networks was implemented for wood identification. On the application side, a discrimination model based on CT images that can be used for nondestructive diagnosis of cultural property wood products was constructed.
|
| Free Research Field |
森林科学
|
