apidays Munich 2025 - Geospatial Artificial Intelligence (GeoAI) with OGC APIs, Emmanuel Mondon (AdviceGEO)

Geospatial Artificial Intelligence (GeoAI) with OGC APIs
Emmanuel Mondon, Independent Consultant at AdviceGEO

apidays Munich 2025 - Accelerate AI Use Cases with APIs
July 2 & 3, 2025

------

Check out our conferences at https://www.apidays.global/

Do you want to sponsor or talk at one of our conferences?
https://apidays.typeform.com/to/ILJeAaV8

Learn more on APIscene, the global media made by the community for the community:
https://www.apiscene.io

Transcript

1. Geospatial Artificial Intelligence (GeoAI) with OGC APIs 3 July 2025 Apidays Munich 2025 Emmanuel Mondon Independent Consultant AdviceGEO
2. About Geospatial
3. Definition of geospatial from the Cambridge Business English Dictionary © Cambridge University Press https://dictionary.cambridge.org/dictionary/english/geospatial Geospatial data means any data that describes something based on its location Geospatial data is also known as location-based or spatial data https://1spatial.com/news/what-is-geospatial-data-and-how-do-you-use-it-2022/ Location i.e. (X, Y, Z, t)
4. Geospatial Data (or geoinformation or georeferenced) Data generated by space assets SatNav SatEO SatCom Data generated by ground assets Data generated by aerial assets Geodesy - Location relative to Earth - thematic characteristics Photogrammetry IoT Personal data i.e. GDPR
5. High-Value Datasets (HVD) According to the Directive on open data and the re-use of public sector information (PSI), ‘high-value’ means data with the potential to: • generate significant socio-economic or environmental benefits and innovative services; • benefit a high number of users, in particular small to medium sized enterprises (SMEs); • assist in generating revenues; and • be combined with other datasets. Six thematic categories of high-value datasets are identified in the Directive: • Geospatial • Earth observation and environment • Meteorological • Statistics • Companies and company ownership • Mobility https://eurogeographics.org/news/high-value-datasets/ https://eurogeographics.org/news/high-value-datasets/ https://eur-lex.europa.eu/legal-content/EN/TXT/?qid=1561563110433&uri=CELEX:32019L1024
7. CONFIDENTIAL & RESTRICTED Examples of Domain Data Spaces Digital Infrastructure Location Ecosystem Connector / Agent bringing interoperability across Data Spaces : Simpl, FIWARE NGSI API, Eclipse Dataspace Connector, etc…
8. Simpl: Cloud-to-edge federations empowering EU data spaces https://digital-strategy.ec.europa.eu/en/policies/simpl • Simpl is an open source, smart and secure middleware platform that supports data access and interoperability among European data spaces • Simpl plays a major role in the creation of the Common European Data Spaces
9. About Open Geospatial Consortium (OGC) https://www.ogc.org/ • Created in 1994 • more than 450 organizations strong—united by a shared passion for geospatial innovation • a dynamic community of innovators from government, business, research institutions, startups • from different sectors, countries and backgrounds across the geospatial world • For 30+ years, the place where collaboration and experimentation come to life • It’s not just about technology • it’s about people working together to make something bigger than any one organization • Collaborative Solutions and Innovation Program (COSI) brings OGC members together to engage in pilots and testbeds
10. Training Data Markup Language for Artificial Intelligence (TrainingDML-AI) https://www.ogc.org/standards/trainingdml-ai • Training data plays a fundamental role in Earth Observation (EO) Artificial Intelligence Machine Learning (AI/ML), especially Deep Learning (DL) • Standardizing any training data used to train, validate, and test AI/ML models that involve location or time • Part 1 approved as official OGC Standard in September 2023 • Standard provides detailed metadata for formalizing the information model of training data • TrainingDML-AI Standard aims to develop the UML model (Conceptual Model) and encodings (JSON and XML) for geospatial machine learning training data
12. Geospatial data is shaping the future: Xoople https://www.xoople.com/ A new paradigm of Earth Data for Enterprise Our transformative end-to-end system has been developed together with customers in multiple industries and geographies to solve real-world needs, to deliver a consistent, high-quality, replicable, AI-ready data set that facilitates solving real-world problems. It will produce a stream of Earth Data designed to be processed by powerful AI at scale, integrating with Geospatial Reasoning and Enterprise Intelligence to recognize patterns, detect change and provide predictive insights about our world.
13. Geospatial Al Platforms: FlyPix AI (DE) & Picterra (CH)
14. Building Blocks for Location Information https://ogcapi.ogc.org/
15. OGC APIs https://www.ogc.org/standards/
16. OGC API - Discrete Global Grid Systems (DGGS) https://ogcapi.ogc.org/dggs/ https://ogcapi.ogc.org/dggs/overview.html • OGC API - DGGS specifies an API for accessing data organised according to Discrete Global Grid Reference Systems (DGGRS) • A DGGRS is a spatial reference system combining • a discrete global grid hierarchy (DGGH, a hierarchical tessellation of the globe into zones at successive refinement levels) • with a zone identifier reference system (ZIRS) to uniquely address these zones • Additionally, to enable DGGS-optimized data encodings, a DGGRS defines a deterministic order for sub-zones whose geometry is at least partially contained within a parent zone of a lower refinement level • A Discrete Global Grid System (DGGS) is an integrated system implementing one or more DGGRS together with functionality for quantization, zonal query, and interoperability • DGGS are characterized by the properties of the zone structure of their DGGHs, geo-encoding, quantization strategy and associated mathematical functions
20. About Discrete Global Grid Systems (1/2) • Artificial Intelligence (AI) is only as effective as the data it processes. While AI models are advancing rapidly, their true potential is limited by the structure, quality, and interoperability of data • Discrete Global Grid Systems (DGGS) provide a spatial data framework that serves as a precursor and efficient tensor lake for AI models • By structuring data into hierarchical, uniform zones, DGGS ensures that AI models can seamlessly integrate, analyze, and scale across diverse geospatial datasets without costly reprocessing • Discrete Global Grid Systems for Artificial Intelligence (DGGS4AI) highlights how DGGS enhances AI capabilities through structured data, interoperability, and efficient spatial representation • DGGS is a key enabler for advanced spatial AI applications • Digital-native, zone-based geospatial representation is crucial, not only for answering complex geographic questions that rely on multi-source data and spatial analysis, but also for building interoperable digital ecosystems across domains
21. About Discrete Global Grid Systems (2/2) • DGGS represent locations as cells, moving beyond traditional geographic reference systems • DGGS can provide a foundation for a global localization system in which every object in the real world has a unique location identifier • DGGS are well-suited for data integration and efficient querying and analysis • DGGS are ideally positioned to serve as authoritative ‘content stores’ for AI-powered natural language queries
22. OGC AI-DGGS for Disaster Management Pilot
23. Next step at apidays Paris in December 2025