Describing ML Models with the Geospatial Machine Learning Model Catalog (GMLMC).
During the height of the COVID-19 pandemic, the government of Togo launched a program to “boost national food production in response to the COVID-19 crisis by distributing aid to farmers”1. To accomplish this, the government needed accurate information about the distribution of smallholder farmers throughout the country. This kind of cropland map did not exist for the country, so they worked with NASA Harvest to rapidly develop a cropland map using AI. Finding enough high-resolution labeled training data to train the machine learning model was also a significant challenge, so the team combined global and local crowdsourced labels collected using the Geo-Wiki platform2 with hand-labeled imagery in targeted areas to train a new model for predicting crop areas.
Describing ML Models with the Geospatial Machine Learning Model Catalog (GMLMC).
Building machine learning models with open training data for precision agriculture and flood detection in Rwanda.
Our Community Voice for this quarter is Gedeon Muhawenayo, a machine learning research engineer at the Rwanda Space Agency working on machine learning for satellite and aerial image processing.Gedeon is an avid user of the open machine learning training datasets available on Radiant MLHub. In this Q&A, Gedeon talks to us about building machine learning models for precision agriculture and flood detection in Rwanda.
Now featuring 250+ organizations that focus on machine learning applications with satellite data
The latest interactive Machine Learning for Earth Observation Market Map, a curated list of organizations focused on different machine learning aspects with a satellite data pipeline, is available for download. This release includes an additional list of 100+ organizations, thanks to a crowdsourcing effort on social media. Earlier in September, we asked our followers on Twitter and LinkedIn to identify organizations that we missed in the earlier version of the market map or were established since then. The large number of contributions from people in such a short period speaks of the niche area of machine learning (ML) for Earth observation (EO). The entries hint toward the incredible aptitude of organizations to optimize these innovative technologies and expand them in the service of humanity.
Following up on our SpatioTemporal Asset Catalog 1.0.0 announcement, this post will finish our mini–series diving into the STAC specification and the ecosystem around it. This one is really about the future, where we see STAC heading in the next six months and beyond. And after this I hope to start an in-depth series of STAC posts that dives deeper into individual projects, highlighting all the great software and data in the STAC ecosystem.
The ‘STAC Ecosystem 1.0’ Vision
One of the core tenents of the STAC Community is that we focus on the process of building an interoperable ecosystem, with the specification serving as a record of the current state of collaboration. With the release of 1.0.0, it is our hope that the core of that collaboration now has an incredibly solid foundation. But there is much, much more to do in order to realize the vision of STAC: we won’t be ‘done’ until every single ‘geospatial asset’ has an associated STAC record.
The Locus Charter promotes responsible practice in the use of location data across all sectors.
We are pleased to announce that we’ve joined the Locus Charter, together with the American Geographical Society (AGS), Association for Geographical Information (AGI), Environmental Information Systems Africa (EIS — Africa), the National Institute of Statistics and Geography (INEGI — Mexico), Open Geospatial Consortium (OGC), PLACE, Royal Geographical Society (RGS) and the Royal Institution of Chartered Surveyors (RICS).
The Locus Charter is a collection of ten international principles to support the ethical and responsible practice of individuals and organizations when using location data.
Building application-ready tools and data for policymakers, resource managers, and other scientists to understand global dynamics in human-environment systems.
Our Community Voices for this quarter are Dr. Carsten Meyer, Mr. Ruben Remelgado, Dr. Steffen Ehrmann, and Ms. Caterina Barrasso from the German Centre for Integrative Biodiversity Research (iDiv) Macroecology and Society Lab. They are working on several projects to detect and understand global dynamics in human-environment systems, focusing on human land use, its underlying societal drivers, and its ecological consequences. The research team uses numerous datasets from Radiant MLHub to model crop suitability layers, which will inform the systematic downscaling of crop statistics into pixel-scale crop type classifications.
Using STAC to catalog machine learning training data.
Researchers and data scientists are increasingly combining Earth observation (EO) with ground truth data from a variety of sources to build faster, more accurate machine learning (ML) models to gain valuable insights in domains ranging from agriculture to autonomous navigation to ecosystem health monitoring. These models are integrated into analytic pipelines that generate on-the-fly predictions at scale. The accuracy of these inferences are then evaluated using well-defined validation metrics and the results used to improve the performance of the original model in a continuous feedback loop.
If this sounds like a complex process, that’s because it is! Ad-hoc techniques for handling these workflows may work well within a single organization, but can lead to a bewildering array of algorithms and data for end-users.
The SpatioTemporal Asset Catalog (STAC) Community is pleased to announce the release of version 1.0.0-beta.2 of the STAC API specification! A big thanks to Phil Varner for leading this release, and to everyone else who pitched in.
What is STAC API? As we are welcoming many new people to the STAC community with the core STAC 1.0.0 release it’s probably worth explaining what this ‘API’ release is all about. STAC originally started from a desire to make a common API to help interoperability between satellite data providers, but soon evolved to focusing on the core JSON language to enable more general geospatial interoperability. The STAC repository initially contained both the API specification along with the three ‘core’ specs (Item, Catalog & Collection). But it was clear that the API really depends on the core, and expands it with additional functionality, so after version 0.9.0 we decided to split STAC into two repositories.
Designed to strengthen practitioners’ local capacity and skills in support of creating impactful machine learning applications.
We have the pleasure of introducing the online Machine Learning for Earth Observations (ML4EO) course. ML4EO technologies present a game-changing opportunity to identify and address unique, complex, and emerging challenges at local, regional, and global scales more accurately and more quickly. The plethora of EO data, combined with ML techniques, can help humanity see, understand, and respond more effectively to a rapidly changing world. But data scientists specializing in ML for EO are underrepresented, especially in Africa. Therefore, in May, we partnered with the Ugandan AI and data science research group at Makerere University to train a community of African regional experts collaborating on solutions to the continent’s needs and problems using ML4EO technologies.
This post continues the STAC 1.0.0 announcement and updates on the community and spec, with an in-depth dive into one part of what we call the ‘STAC Ecosystem’. This is the ever-growing set of libraries, clients, and servers that rely on the STAC specification to build towards our goal of interoperability. The advanced state of this ecosystem is one of the things I’m most proud of with STAC, as one of the earliest goals was to have as much real-world validation as possible. And the coolest thing is that it feels like the momentum is just starting to really hit an inflection point, with more and more tools supporting STAC.
So this post will give an overview of the current state of STAC-related software, and where our ‘STAC 1.0.0 Initiative’ sponsor money has enabled us to accelerate. And in the next post, I’ll share what else we’re planning to fund, towards a vision of ‘STAC Ecosystem 1.0.0’. That’s the point when there are complete tools, with all the appropriate tutorials and documentation, so that anyone can easily create or consume STAC. Our goal is that most people will be able to do so without having to ever read the specification. It will just be the foundation that enables a network of interoperable data.