This month features a variety of podcasts including the importance of labels, flood monitoring, Microsoft’s Planetary Computer, and more. Catch up on webinars from FOSS4G 2021, Geo For Good, PhiWeek, and FedGeoDay. Explore data challenges, the latest ML methodologies and applications, and other helpful resources. Explore the highlights from November:
Methodologies & Applications
- Deep learning methods for daily wildfire danger forecasting
- Mapping access to water and sanitation in Colombia using publicly accessible satellite imagery,crowd-sourced geospatial information and random forests
- Disaster mapping from satellites: Damage detection with crowdsourced point labels
- An analysis of elephants’ movement data in Sub-Saharan Africa using clustering
- Remote sensing image super-resolution and object detection: Benchmark and state of the art
- Flood damage assessment with Sentinel-1 and Sentinel-2 data after Sardoba dam break with GLCM features and Random Forest method
- Expanding infrastructure and growing anthropogenic impacts along Arctic coasts using Sentinel data and machine learning approach
- Automated object detection on aerial images for limited capacity embedded device using a lightweight CNN model
- Understanding the requirements for surveys to support satellite-based crop type mapping: evidence from Sub-Saharan Africa
- Quick detection of field-scale soil comprehensive attributes using UAV and Sentinel remote sensing data via the integration of UAV and Sentinel-2B remote sensing with a random forest
- SustainBench: Benchmarks for monitoring the Sustainable Development Goals with machine learning
- TorchGeo: Deep learning with geospatial data
- The Swiss data cube, analysis ready data archive using Earth observations of Switzerland
Data Access & Validation
Standards Research & Innovation
- PySTAC v1.2.0 released with support for Table Extension, Python 3.10, and more
- New JavaScript STAC software/tool “STAC Layer (for Leaflet)”
Tutorials, Webinars (recorded) & Resources
TUTORIALS
- Example usage of TorchGeo using the Tropical Cyclone Wind Estimation Competition dataset available on RadiantMLHub with PyTorch Lightning
- Integrating satellite image analysis into urban planning GIS
- Area monitoring: Detecting pixel-level mowing marker when an object-based approach is not good enough
- Data-fusion sand-oriented land cover classification: Modified Normalized Difference Sand Index (MNDSI) with Sentinel Hub
- How to access Sentinel-2 data from AWS and find STAC items to load
- How to transform geospatial data to Cloud-Native Frameworks with Element 84 on AWS
- ISPRS SC Summer School: Deep learning for remote sensing images with R language
WEBINARS
- FOSS4G: Open AI data to address environmental sustainability challenges with Hamed Alemohammad and Rob Emanuele
- FOSS4G: Getting started with STAC for public datasets with Matthew Hanson
- FOSS4G: The State of STAC with Matthew Hanson
- FOSS4G: Mapping floods in urban areas from space at local risk level with Guy Schumann
- FOSS4G: SatProc, an open-source library to train and deploy Deep Segmentation Neural Nets for geospatial imagery with Maria Devsa and Damian Silvani
- QGIS: Accessing Digital Earth Africa datasets using QGIS with Edward Boamah
- ESA Phiweek 202 side events playlist
- FedGeoDay 2021 presentations playlist
- Geo for Good Summit 2021 playlist
PODCASTS
- Mapscaping: Mallory Dodd on Why quality training data and labels matter
- Mapscaping: The Planetary Computer with Rob Emanuele
- Mapscaping: Flood monitoring from space with Shay Strong
- Mapscaping: Collecting and processing aerial image at scale with Mike Bewley
- Down to Earth: Jón Benediktsson, Paolo Gamba, and Naoto Yokoya discuss the evolution of machine learning in Earth observation
- Down to Earth: Manil Maskey on maintaining scientific integrity in remote sensing
- Down to Earth: Beth Tellman on combating climate change with data
- Africa GeoConvo: The ESA open mapping hub with Monica Nthiga
- The Scene from Above: The latest EO news
RESOURCES
- Radiant Earth Foundation and RCMRD to develop joint capacity building program focused on machine learning for Earth observation in Africa
- Introducing a novel very high resolution dataset of landfills and waste dumps
- Highlights from Land Processes Distributed Active Archive Center, data available for modelling of changes and assessment of landscape data
- Lessons learned in teaching machine learning for Earth observation techniques after the training of trainers bootcamp
- Application for ClimateChange AI summer school to tackle climate problems with AI due Dec. 17
- Cambridge University Press call for papers on environmental informatics
- Sentinel Hub provides geotagged photo app to communicate with farmers
- Google AI expands machine learning-based flood forecasting system
- The development of the World Settlement Footprint Suite, the world’s most comprehensive dataset on human settlement
Data Challenges
- AI4FoodSecurity Challenge, South Africa and Germany (ends Dec 19)
- Microsoft AI For Earth: On Cloud N: Cloud Cover Detection Challenge (ends Feb 7 2022)
- Omdena – Using AI to Identify Optimal Locations Spots for Floating Solar Installations (ends Feb 2022)
- Omdena – Drone Data and Environmental Sensor Read-out to Automate Plant Health Prediction and Drought Control (ends Feb 2022)
- EarthNet 2021: A machine learning challenge for Earth surface and localized impact forecasting
Conferences & Workshops
- GMES and Africa (Dec 6-10)
- AGU Fall Meeting 2021 (Dec 13-17)
- NeurIPS 2021 Workshop (Dec 14)
- IARAI Workshop on Complex Data Challenges in Earth Observation (Jan 11)
- ESIPFed January meeting (Jan 18-21)
What are we missing? Contact Louisa@radiant.earth