• Search Research Projects
  • Search Researchers
  • How to Use
  1. Back to project page

2021 Fiscal Year Research-status Report

Post-disaster Recovery Monitoring based on Multi-Source Remote Sensing Imagery and Deep Learning

Research Project

Project/Area Number 21K14261
Research InstitutionThe University of Tokyo

Principal Investigator

郭 直霊  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
KeywordsPost-disaster Monitoring / Change Detection / Deep Learning / Remote Sensing
Outline of Annual Research Achievements

First, multi-source remote sensing imagery with related ground truth, including semantic segmentation annotation, object bounding box, and DEM are collected.
Then, weak supervision based multi-task deep learning methods are formulated to train the urban mapping model, which can conduct semantic segmentation, object detection, and DEM generalization simultaneously even with the inexact and inadequate training dataset.
Meanwhile, a learnable weighting method for multi-loss functions combination is proposed to achieve automatic fine-tuning.
The multi-task urban mapping will serve the following steps with a high-performance multi-task mapping model, and provide additional land cover mapping results as well.

Current Status of Research Progress
Current Status of Research Progress

3: Progress in research has been slightly delayed.

Reason

The quality of DEM dataset is not good enough.
The learnable weighting method for multi-loss functions is still under development.

Strategy for Future Research Activity

Post-disaster Change Detection will be the finished in FY2022.
First, multi-temporal post-disaster remote sensing imagery and the ground-truth of changing will be collected. Then, with the help of (A) Multi-task Urban Mapping, the land cover semantic segmentation, object detection, and DSM in multi-temporal can be generated. After that, an end-to-end deep learning model for change detection will be trained by data fusing and ensemble learning. Meanwhile, to solve the slight misalignment of multi-temporal imagery as well as the imbalanced land cover ratio, a specific loss function will be proposed. Finally, the trained change detection model will be capable of detecting land cover changes, such as the destroyed, new vacant, under construction, completed, etc.

Causes of Carryover

The research need to prepare UAV as well as GPU server; 1.Unmanned Aerial Vehicle (UAV)(DJI, Phantom 4 Pro V2.0); 2.NVIDIA GeForce RTX 3090 24GB GDDR6X.
Join the international conference for sharing research results.
Hire part-time students for the experiment.

  • Research Products

    (4 results)

All 2022 2021

All Journal Article (4 results) (of which Int'l Joint Research: 3 results,  Peer Reviewed: 3 results,  Open Access: 2 results)

  • [Journal Article] Understanding the Urban Environment from Satellite Images with New Classification Method: Focusing on Formality and Informality2022

    • Author(s)
      Cheng Qianwei、Zaber Moinul、Rahman AKM Mahbubur、Zhang Haoran、Guo Zhiling、Okabe Akiko、Shibasaki Ryosuke
    • Journal Title

      Sustainability

      Volume: 14 Pages: 4336~4336

    • DOI

      10.3390/su14074336

  • [Journal Article] Benchmark Analysis for Robustness of Multi-Scale Urban Road Networks Under Global Disruptions2022

    • Author(s)
      Shang Wen-Long、Gao Ziyou、Daina Nicolo、Zhang Haoran、Long Yin、Guo Zhiling、Ochieng Washington Y.
    • Journal Title

      IEEE Transactions on Intelligent Transportation Systems

      Volume: 1 Pages: 1~11

    • DOI

      10.1109/TITS.2022.3149969

    • Peer Reviewed / Int'l Joint Research
  • [Journal Article] Understanding rooftop PV panel semantic segmentation of satellite and aerial images for better using machine learning2021

    • Author(s)
      Li Peiran、Zhang Haoran、Guo Zhiling、Lyu Suxing、Chen Jinyu、Li Wenjing、Song Xuan、Shibasaki Ryosuke、Yan Jinyue
    • Journal Title

      Advances in Applied Energy

      Volume: 4 Pages: 100057~100057

    • DOI

      10.1016/j.adapen.2021.100057

    • Peer Reviewed / Open Access / Int'l Joint Research
  • [Journal Article] GRAPH NEURAL NETWORK BASED MULTI-FEATURE FUSION FOR BUILDING CHANGE DETECTION2021

    • Author(s)
      Yuan W.、Yuan X.、Fan Z.、Guo Z.、Shi X.、Gong J.、Shibasaki R.
    • Journal Title

      The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences

      Volume: XLIII-B3-2021 Pages: 377~382

    • DOI

      10.5194/isprs-archives-XLIII-B3-2021-377-2021

    • Peer Reviewed / Open Access / Int'l Joint Research

URL: 

Published: 2022-12-28  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi