2018 Fiscal Year Final Research Report
Construction of vector data cloud connecting plant structure and numerical information
| Project/Area Number |
16K16043
|
|---|
| Research Category |
Grant-in-Aid for Young Scientists (B)
|
| Allocation Type | Multi-year Fund |
|---|
| Research Field |
Information network
|
|---|
| Research Institution | Shinshu University |
|---|
Principal Investigator |
|
|---|
| Research Collaborator |
Genno Hirokazu
|
|---|
| Project Period (FY) |
2016-04-01 – 2019-03-31
|
| Keywords | 農園モニタリング / フィールドモニタリング / 農業情報 / 深層学習 / 訓練データ拡張 / 植物3次元構造復元 / スマート農業 |
|---|
| Outline of Final Research Achievements |
In this research, we developed the automatic extraction of the numerical information such as fruit shape information by using deep learning with collected field images from monitoring systems. We developed a monorail typed mobile field monitoring system to reconstruct 3D data of the field and to extract plant height numerical information. We also developed a deep learning training technology to automatically extract the positions and their shape information of fruits from a field monitoring image. Although deep learning generally requires a lot of time and effort to collect training data, our technology significantly reduces the time and effort by automatically generating annotated artificial field images.
|
| Free Research Field |
スマート農業
|
| Academic Significance and Societal Importance of the Research Achievements |
本研究で開発した手法で数値化された植物構造情報を抽出することにより，作物の成長予測シミュレーションや予測モデルの構築に活用できる．単視点画像であっても数値化された植物情報を抽出する技術も開発したため，過去の農園画像に対しても利用が可能である．また，観察データから重要な遺伝子の働きを特定するフェノミクス研究において，遺伝子型と表現型との対応付けの効率化に貢献できる．
|
