| Project/Area Number |
21K14261
|
|---|
| Research Category |
Grant-in-Aid for Early-Career Scientists
|
| Allocation Type | Multi-year Fund |
|---|
| Review Section |
Basic Section 22050:Civil engineering plan and transportation engineering-related
|
|---|
| Research Institution | The University of Tokyo |
|---|
Principal Investigator |
Guo Zhiling 東京大学, 空間情報科学研究センター, 客員研究員 (40897716)
|
|---|
| Project Period (FY) |
2021-04-01 – 2024-03-31
|
| Project Status |
Completed (Fiscal Year 2023)
|
| Budget Amount *help |
¥4,680,000 (Direct Cost: ¥3,600,000、Indirect Cost: ¥1,080,000)
Fiscal Year 2023: ¥780,000 (Direct Cost: ¥600,000、Indirect Cost: ¥180,000)
Fiscal Year 2022: ¥1,950,000 (Direct Cost: ¥1,500,000、Indirect Cost: ¥450,000)
Fiscal Year 2021: ¥1,950,000 (Direct Cost: ¥1,500,000、Indirect Cost: ¥450,000)
|
|---|
| Keywords | Deep Learning / Geoscience / Post-disaster Monitoring / Remote Sensing / Computer Vision / Change Detection |
|---|
| Outline of Research at the Start |
This research addresses the challenge of achieving high performance recovery monitoring and related tasks based on inexact and inadequate training dataset. The weak-supervised learning, multi-task learning, transfer learning is proposed.
This research sets up an architecture which makes connecting between the research fields of urban mapping, change detection, and recovery monitoring.
Based on this research, more interdisciplinary datasets and methods are expected to for post-disaster recovery monitoring optimization to promote the efficiency and accuracy.
|
| Outline of Final Research Achievements |
Improved monitoring of post-disaster recovery is important, especially in Japan where natural disasters cause heavy losses frequently. Recently, machine learning and remote sensing data have been successfully applied in different aspects of disaster risk management. However, the monitoring in the recovery phase has not been paid sufficient attention by empirical studies, and the high-performance monitoring remains a formidable challenge. Focusing on the perspective of physical recovery, this research proposed to achieve accurate, rapid, and cost-effective post-disaster recovery monitoring based on multi-source remote sensing imagery and deep learning methods. Three stages are composed in the proposed method: 1) weak supervision for multi-task urban mapping, 2) end-to-end change detection via data fusion, 3) transfer learning based recovery monitoring. The method of this research attempts to promote the capability of pre-disaster recovery planning, evaluation, and management.
|
| Academic Significance and Societal Importance of the Research Achievements |
この研究は、日本における災害後の復旧監視を向上させるため、多元なリモートセンシングとディープラーニングを応用して正確で迅速かつ費用効果の高い解決策を提供します。新しい3段階の手法を用いて、復旧フェーズに取り組み、災害復興計画と管理を改善します。
|
