Data-driven, dynamical complex systems models can enhance disaster resilience by capturing dynamic interdependencies and intrinsic resilience characteristics of communities.

Toward data-driven, dynamical complex systems approaches to disaster resilience

Emerging technologies can enable low-cost satellite-based systems for real-time disaster monitoring and relief assistance, providing high-resolution imagery from geosynchronous and L1 altitudes.

The role of emerging technologies in imagery for disaster monitoring and disaster relief assistance

Night-time light remote sensing images can effectively measure community resilience in natural disasters, providing actionable information for developing resilient and sustainable communities.

Observing community resilience from space: Using nighttime lights to model economic disturbance and recovery pattern in natural disaster

CNN-based automatic natural disaster detection from aerial imagery accurately detects landslides and floods with 80%-90% accuracy, aiding in disaster relief efforts.

Disaster detection from aerial imagery with convolutional neural network

OmbriaNet, a deep neural network architecture, effectively detects changes between permanent and flooded water areas using multitemporal Sentinel-1 and Sentinel-2 data fusion, producing high-quality flood maps.

OmbriaNet—Supervised Flood Mapping via Convolutional Neural Networks Using Multitemporal Sentinel-1 and Sentinel-2 Data Fusion

Big data technologies can enhance disaster resilience by improving information linkages and enhancing socio-ecological vulnerability management.

Disaster resilience through big data: Way to environmental sustainability

Nighttime lights remote sensing images can provide valuable insights into community recovery and resilience in natural disasters, improving disaster awareness and preparedness.

Spatial Assessment of Community Resilience from 2012 Hurricane Sandy Using Nighttime Light

The proposed framework using deep learning and open data effectively segmented aerial images for disaster impact assessment, reducing manual annotation and requiring less parameters compared to state-of-the-art networks.

Deep Learning-based Aerial Image Segmentation with Open Data for Disaster Impact Assessment

High-resolution, high-caching satellite imagery from Planet Labs' constellation of satellites enhances flood risk forecasting, historical modeling, and risk management.

Flooding applications enabled by high resolution, high cadence imagery from the Planet constellation of satellites

Lightweight real-time semantic segmentation models show potential for efficient and accurate aerial image segmentation during flooding events, with some achieving over 60% test mIoU on the FloodNet dataset.

Comparative Study of Real-Time Semantic Segmentation Networks in Aerial Images During Flooding Events

A hybrid crowdsourcing and machine learning solution can rapidly process large volumes of aerial data for disaster response, aiding in wildlife protection, human rights, and archeological exploration.

Combining Human Computing and Machine Learning to Make Sense of Big (Aerial) Data for Disaster Response

Our CNN-based framework effectively detects change in satellite images, identifying the most severely affected areas in disasters, improving disaster mapping efficiency and reducing resource-related errors.

From Satellite Imagery to Disaster Insights

CNN models trained on Volan2018 dataset effectively detect critical ground assets like building roofs, vehicles, vegetation, debris, and flooded areas in aerial images, aiding disaster response and recovery.

Convolutional neural networks for object detection in aerial imagery for disaster response and recovery

Swift trust, information sharing, and public-private partnerships are critical enablers of resilience in disaster supply chain networks, as demonstrated in the Nepal earthquake aftermath.

The role of Big Data in explaining disaster resilience in supply chains for sustainability

The proposed method improves spatial resolution of satellite-based SAR imagery, enabling early detection of small-scale disasters and preventing secondary occurrence of large-scale disasters.

Method for Frequent High Resolution of Optical Sensor Image Acquisition using Satellite-Based SAR Image for Disaster Mitigation

FloodNet, a high-resolution UAV imagery dataset, can improve post-disaster damage assessment by detecting flooded roads, buildings, and distinguishing between natural and flooded water.

FloodNet: A High Resolution Aerial Imagery Dataset for Post Flood Scene Understanding

The proposed disaster resilient smart city architecture, integrating IoT and big data analytics, effectively detects and manages disasters, demonstrating superior performance in processing time and throughput.

Towards Disaster Resilient Smart Cities: Can Internet of Things and Big Data Analytics Be the Game Changers?

The lightweight transfer learning-based model effectively assesses building conditions in aerial images, reducing memory and computation time while maintaining accuracy.

A Lightweight Transfer Learning-Based Model for Building Classification in Aerial Imagery

Big data can enhance disaster resilience by providing early warning systems, weather forecasting, and immediate responses, but challenges in data collection, analytics, infrastructure, and ethics must be addressed for maximum benefits.

Big data-driven disaster resilience

Image4Act effectively processes social media imagery during disasters, aiding humanitarian organizations in situational awareness and launching relief operations.

Image4Act: Online Social Media Image Processing for Disaster Response

planet data skynet imagery and disaster resilience

These studies suggest that data-driven models, satellite-based systems, night-time light remote sensing, and CNN-based detection can collectively enhance disaster resilience through dynamic modeling, real-time monitoring, resilience measurement, and accurate disaster detection.