/Natural-Disaster-Damage-Assessment-Deep-Learning

This repo contains datasets, papers and other information related to Destruction Detection in Satellite Imagery.

Natural Disaster Damage Assessment

The datasets for damage assessments are divided into the following categories:

  1. Non-Imaging Data (Text, Tweets, Social Media Post)
  2. Imaging Dataset:
    1. Ground Level Images
    2. Aerial Imagery (UAV)
    3. Satellite Imagery

Datasets

  1. xView, 2018 | Satellite
  2. xView2, 2020 | Satellite
  3. AIDER, 2020 | UAV
  4. ISBDA, 2020 | UAV
  5. Syria Destruction Dataset, 2021 | Satellite
  6. LIVER-CD, 2021 | Satellite
  7. FloodNet, 2021 | UAV
  8. Ida-BD: Hurricane Ida, 2023 | Satellite

Papers

2019

  1. Building Damage Detection in Satellite Imagery Using Convolutional Neural Networks, 2019 | Paper

2020

  1. An Attention-Based System for Damage Assessment Using Satellite Imagery, 2020 | Paper
  2. Assessing Post-Disaster Damage from Satellite Imagery using Semi-Supervised Learning Techniques, 2020 | Paper
  3. BUILDING DISASTER DAMAGE ASSESSMENT IN SATELLITE IMAGERY WITH MULTI-TEMPORAL FUSION, 2020 | Paper
  4. Cross-directional Feature Fusion Network for Building Damage Assessment from Satellite Imagery, 2020 | Paper
  5. Destruction from sky: weakly supervised approach for destruction detection in satllite imagery, 2020 | Paper
  6. FloodNet: A High Resolution Aerial Imagery Dataset for Post Flood Scene Understanding, 2020 | Paper
  7. RescueNet: Joint Building Segmentation and Damage Assessment from Satellite Imagery, 2020 | Paper

2021

  1. Building Damage Detection Using U-Net with Attention Mechanism from Pre- and Post-Disaster Remote Sensing Datasets, 2021 | Paper
  2. Weakly Supervised Segmentation of Small Buildings with Point Labels, 2021 | Paper

2022

  1. Hybrid U-Net: Semantic segmentation of high-resolution satellite images to detect war destruction, 2022 | Paper
  2. Interpretability in Convolutional Neural Networks for Building Damage Classification in Satellite Imagery, 2022 | Paper
  3. Self-Supervised Learning for Building Damage Assessment from Large-scale xBD Satellite Imagery Benchmark Datasets, 2022 | Paper
  4. SegDetector: A Deep Learning Model for Detecting Small and Overlapping Damaged Buildings in Satellite Images, 2022 | Paper

2023

  1. LARGE-SCALE BUILDING DAMAGE ASSESSMENT USING A NOVEL HIERARCHICAL TRANSFORMER ARCHITECTURE ON SATELLITE IMAGES, 2023 | Paper
  2. xFBD: Focused Building Damage Dataset and Analysis, 2023 | Paper
  3. RescueNet: A High Resolution UAV Semantic Segmentation Dataset for Natural Disaster Damage Assessment, 2023 | Paper | Code

Detection Papers

  1. CVNet: Contour Vibration Network for Building Extraction, 2022 | Paper
  2. PolyWorld: Polygonal Building Extraction with Graph Neural Networks in Satellite Images PP-LinkNet: Improving Semantic Segmentation of High Resolution Satellite Imagery with Multi-stage Training.pdf Sat2Graph: Road Graph Extraction through Graph-Tensor Encoding.pdf

Others

  1. SUSTAIN BENCH : Benchmarks for Monitoring the Sustainable Development Goals with Machine Learning, 2021 | Paper
  2. SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers, 2021 | Paper | Code