When AI meets GIS: UN Global Pulse and FAO co-chair UN Geo-AI Working Group

5 min read
Share on facebook
Share on twitter
Share on linkedin
Share on email

The Geo-AI Working Group, co-chaired by the Food and Agriculture Organization of the UN and UN Global Pulse, used its first year to share strategies and expertise between UN agencies, academics, and industry professionals in order to envision how AI can best be leveraged in the world of geographic information systems (GIS). The Group is part of the UN Open GIS Initiative. 

What is UN Open GIS?

The UN Open GIS Initiative, established in March 2016, is an ongoing initiative of the United Nations Department of Operational Support. Officially a Partnership Initiative for Technology in Peacekeeping, it was created to develop open-source technology in the field of geographic information systems. UN Open GIS is focused on creating a bundle of tools that will be freely available for use by the entire UN System, as well as partner organizations from the academic, public, and private sectors.

Mapping is instrumental to any organization that operates on a global scale. In developing a toolkit specifically suited to the UN’s peacebuilding and peacekeeping operations, UN Open GIS represents an integral part to furthering the UN’s mission. The initiative is made up of five groups, each working towards a specific objective that contributes to the initiative’s overarching goal: Hybrid GIS, Capacity Building, Data Collection, Geo-Analysis, and Geo-AI. 

What is the Geo-AI Working Group?

The newest working group of UN Open GIS was created in 2020, in response to a growing need to acknowledge the impact of artificial intelligence on the broader field of technology. The Geo-AI Working Group’s objective is to leverage AI applications in geospatial work, with a particular emphasis on using artificial intelligence to interpret and analyze satellite imagery.

By training technology to look at images and interpret abstract shapes into concrete data, the Geo-AI group is working towards a way to streamline this process. A computer can process data much faster than the human brain; automating the initial data collection allows for analysts to concentrate on other tasks that can’t be performed efficiently by a machine. 

UN Global Pulse and FAO co-chair the working group and are working together with researchers and industry leaders to research, elaborate and adopt innovations, best practices, and recommendations that can support geospatial analysis. 

What has been accomplished so far?

In its first year of activity, the Geo-AI Working Group gathered experts from across the ecosystem and began ideating how to incorporate their common knowledge into GIS work. The Group held its first consultation in September 2020 to consolidate the expertise of its growing members. The first part of the process saw experts sharing the progress of various teams. For example, the UN Office on Drugs and Crime presented their work using satellite imagery to identify and monitor the farming of illicit crops, which is a laborious manual process. The UNODC believes that implementing Geo-AI to automate the identification of illicit crops, like coca or opium poppy, will drastically reduce the amount of time needed for this process, thus making it much more feasible to provide timely and accurate assessments. 

A team at the GeoAI Smarter Map and LiDAR Lab, at Southwest Jiaotong University, presented a cloud-based web app called GeoIME — the Geospatial Infrastructure Management Ecosystem. This technology uses AI to assess the impact of earthquakes on buildings and bridges, estimating the risk of these structures and providing recommendations for structural reinforcements. Using AI to get this information is an efficient way to survey the entire area of an earthquake without needing human resources in affected areas. The team is hoping to scale the tool to deliver similar information after floods and other natural disasters, as well as expand the scope to include roads and other infrastructure elements.

See the Group’s First Consultation Report for a summary of all presented projects. 

What’s next?

The Geo-AI Working Group already has its sights set on its future work. In addition to the presentation of ongoing projects, the September consultation was an opportunity to discuss priorities that lie ahead as well as challenges. 

One of the Group’s objectives will be to address current challenges in implementing AI into geospatial work, like reducing the obstructions of satellite imagery by cloud cover. Because clouds are a frequent obstacle to getting clear images, this is a top priority for geospatial work, whether those images are seen by human eyes or processed by an AI. Gispo, a geospatial consulting group, is working on a programme that will reduce the effect of clouds, by choosing each pixel of a satellite image from the most recent time with no cloud cover. This creates an optimized mosaic of the image, able to show relatively timely images while circumventing the cloud cover issue. This project is in initial stages, but is a promising start to addressing this challenge.

Participants also highlighted the importance of collaboration among various stakeholder groups. 

The World Geospatial Industry Council (WGIC) expressed interest in collaborating with UN Open GIS on the ethical and personal privacy implications of geospatial work. The Kungliga Tekniska Högskolan Royal Institute of Technology (KTH) proposed a collaboration with the UN in order to build a curriculum for development in the Geo-AI field.

The GEO-AI Working Group is committed to developing a system to share resources, technology, and experiences, and expand into new areas of consideration, such as data mining, social media, and predictive analysis. With this roadmap for what comes next for Geo-AI, the Group is set to begin its second year of activity. 

Did you enjoy this blog post? Share it with your networks!

Share on facebook
Share on twitter
Share on linkedin
Share on email

News, thoughts and ideas about big data and AI, data privacy and ethics from across the Pulse Lab Network. Read more on the blog.

Pulse Lab Jakarta

Expanding the Definition of Scaling Up

The “pilot to scale-up model” is a widely used approach when introducing social interventions or new policies. The premise of the model is very logical

Scroll to Top