Digital Blue World Map, Copyright: gonin / 123RF Stock Photo

I spent last week in London at the Open Geospatial Consortium (OGC) October 2023 Open Standards Code Sprint. Or to put it another way, I spent three full days simply coding, one of my favourite ways to pass the time!

OGC is an international membership organisation supporting a community of over five hundred businesses, government agencies, research organizations, and universities. These organisations work together to make geospatial information Findable, Accessible, Interoperable, and Reusable (FAIR), by undertaking collaborative projects on developing and promoting consensus-based standards for sharing information, together with innovative projects demonstrating the potential for geospatial data and information.

The Code Sprint was sponsored by Ordnance Survey, supported by the European Union Satellite Centre (SatCen), the United States National Geospatial-intelligence Agency (NGA), & the UK’s Defence Science and Technology Laboratory (Dstl). The event focused on applying a variety of OGC standards and specifications to various geospatial datasets.

During the three days I joined several breakouts to improve my knowledge of standards. Pixalytics has recently being working on the OGC Disaster Pilot 2023 Innovation programme, and so I was interested in learning more about the Model for Underground Data Definition and Interchange (MUDDI) as other participants in the Disaster Pilot work had discussed the model in terms of responding to disasters. MUDDI serves as a framework to make datasets for underground objects more easily shared, managed and used. It has many applications, such as improving knowledge when excavating locations to lay/update/fix infrastructure such as pipelines and cables.

In the UK, the MUDDI conceptual model has been used to create a profile for excavation that, in turn, has been the input for the UK’s National Underground Asset Register (NUAR) developed by the Geospatial Commission. NUAR aims to capture the first few metres of the underground environment, improving the efficiency and safety of underground works by providing secure access to privately and publicly owned location data about pipes and cables.

I worked with a number of datasets during the event, including testing the NGA’s Geospatial-Intelligence Imagery Media for Intelligence, Surveillance, and Reconnaissance (GIMI) profile. While a huge title, it is essentially a practical example of implementing a standard, to test whether the standard is written in a way that works, or to identify any issues there may be with it. In this case, I tested the implementation of – we apologise for the upcoming long titles – of the following two standards :

• International Standards Organization/International Electrotechnical Commission Base Media File Format (ISOBMFF) for video/audio, and
• High Efficiently Image File Format (HEIF) for still imagery standards.

Finally, I also spent time discussing the recently released OGC conceptual model standard for Training Data Markup Language for Artificial Intelligence (TrainingDML-AI) – again apologies for the long title! This is something that Pixalytics has been part of the team working on this for the last couple of years. The aim of the standard is to provide consistency for how training datasets for machine learning are established and described, in particular, it describes how:

• the training data should be prepared, covering items such as provenance or quality;
• to specify different metadata used for machine learning tasks such as scene/object/pixel levels;
• to describe the high-level training data information model; and
• to use external classification schemes for aspects such as ground truth labelling.

Overall, it was a hard and long three days at the Code Sprint, but great to spend time in the company of coders in person, together with a larger group joining us online. At Pixalytics we fully support OGC’s aims to make geospatial data easier to use, and for this to be promoted to the widest possible audience. We have been pushing this in Earth Observation for a number of years, and it is lovely to be part of a community focused on this.