| Project/Area Number |
22K20595
|
|---|
| Research Category |
Grant-in-Aid for Research Activity Start-up
|
| Allocation Type | Multi-year Fund |
|---|
| Review Section |
0603:Forestry and forest products science, applied aquatic science, and related fields
|
|---|
| Research Institution | Forest Research and Management Organization |
|---|
Principal Investigator |
SCHAEFER Holger 国立研究開発法人森林研究・整備機構, 森林総合研究所, 研究員 (80897231)
|
|---|
| Project Period (FY) |
2022-08-31 – 2024-03-31
|
| Project Status |
Completed (Fiscal Year 2023)
|
| Budget Amount *help |
¥2,470,000 (Direct Cost: ¥1,900,000、Indirect Cost: ¥570,000)
Fiscal Year 2023: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
Fiscal Year 2022: ¥1,430,000 (Direct Cost: ¥1,100,000、Indirect Cost: ¥330,000)
|
|---|
| Keywords | 森林炭素循環 / 土壌微生物 / 菌根菌 / 菌糸 / 土壌断面撮影 / 画像分析 / 人工知能 / 森林物質循環 / 土壌炭素 / 菌糸生産・分解 / mycorrhizal fungi / hyphae / soil imaging / image analysis / artificial intelligence |
|---|
| Outline of Research at the Start |
In forests, mycorrhizal fungi (菌根菌) receive up to 20% of the carbon assimilated by plants. Measuring how fast the hyphae (菌糸) of mycorrhizal fungi grow and decompose in the soil is, hence, important for understanding the forest carbon cycle. A promising method for measuring the amount of hyphae in soil is image analysis. However, manual soil image analysis is time-intensive and prone to human error. In the proposed research, automated soil image analysis using artificial intelligence is developed, to help advance our understanding of how mycorrhizal fungi affect the forest carbon cycle.
|
| Outline of Final Research Achievements |
In this research project, an AI-based method for the automated measurement of the amount of hyphae on soil profile images was developed. For this, high-resolution images of a soil profile were taken with a self-developed imaging device. On part of the taken images, hyphae were marked manually. An AI was, then, trained with the manually marked im-ages to differentiate between hyphal pixels and soil background pixels. By using the trained AI, the time to analyze soil profile images was reduced from several days (manual analysis) to only a few minutes (automated analysis). Further-more, the amount of hyphae on soil profile images was estimated with high accuracy.
|
| Academic Significance and Societal Importance of the Research Achievements |
殆どの樹木は菌根菌と共生し、炭素を糖類として菌根菌に供給するため、森林土壌における菌糸バイオマスの変動を明らかにすることが、森林炭素循環の動態の理解・予測に向けて重要である。本研究では、人工知能を活用して土壌断面画像における菌糸量変動を手動分析より十分に速く、正確に自動測定できる手法を開発した。開発した手法は森林土壌における炭素循環を定量化する今後の研究に役立つと考えられる。
|
