GISCafe Industry Predictions for 2025 – Edgybees

By Shay Har-Noy, Chief Executive Officer, Edgybees 

Shay Har-Noy

The geospatial sector has been experiencing a significant transformation, driven by rapid technological advancements, new business models, and growing global challenges that demand more sophisticated spatial analysis and decision-making tools. While no one can predict the future, we have seen some trends in commercial usage as well as during the armed conflicts around the world that suggest what 2025 may bring. In 2025, we can expect the integration of geospatial technologies in traditional workflows, enhanced data availability, and smarter tools to reshape the industry. These changes will impact everything from agriculture and urban planning to climate monitoring and disaster management.

Here are some of my major trends and predictions for the geospatial sector by 2025:

1. Increased Integration of AI and Machine Learning in Geospatial Analysis

Artificial intelligence will continue to have an impact on data processing where AI models continue to get more sophisticated and make their way as a standard part of the analyst workflow. The sheer volume of geospatial data—generated by satellites, UAS, sensors, and mobile devices—is overwhelming, but AI has shown that it is capable of helping to extract actionable insights from this data at scale.

One of the challenges of AI deployed in the geospatial domain has traditionally been that one needs to have expertise in geospatial data, imagery AND computer vision/AI. In 2025, we can expect to see AI workflows graduating away from being just in the hands of select technical users with expertise in AI and leveraged by a broader range of users throughout the stack and mission.  Indeed, we have seen tools released by GXP, Orbital Insight, MetaVI, Preligens and others not only become increasingly powerful but also incredibly easy to use. For example, the MetaVI application allows a user to train new models to detect objects with the click of a button.

The trend where geospatial users will be leveraging AI algorithms to automate complex spatial analysis, detect patterns, and even predict future trends will continue with increased penetration not just due to increased sophistication of tools but because more people with different skill sets will be able to use them. Rather than replace humans, AI at users fingertips will allow skilled practitioners extraordinary power to leverage data like never before.

2. Expansion of High-Resolution Imagery where multisource data becomes the norm

The imagery market is expected to continue its growth, particularly in high-resolution imagery. In recent years, companies like Planet Labs, and BlackSky have made strides in creating large constellations of small, low Earth orbit (LEO) satellites that capture high-frequency, high-resolution imagery of the planet’s surface.  In addition, we are seeing a proliferation of UAS being leveraged in the field with both commercial companies and government agencies depending on them for both tactical and strategic data collection.

In 2025, we are going to see the highest resolution sensors such as those operated by Maxar and Airbus used interchangeably with higher frequency constellations. Many errors in geospatial imagery analysis can be attributed to sampling error and detection error.  For an extreme example, consider the use case of monitoring a port and trying to identify specific small vessels. If you have a the highest resolution sensor in the world that is only able to image every other day, you may be able to easily identify these small watercrafts in the imagery you get, however, you will only be able to get identifications every other day – this is high sampling error. Alternatively, if you are collecting low resolution imagery multiple times a day, you may not be able to distinguish between the small watercraft in the water – this is high detection error. Neither of which is desirable.

In 2025, we will see the increased use of high resolution AND high frequency by leveraging different satellites from various commercial companies.  Indeed, the EOCL contract from NRO sets us on that direction and we are seeing increased adoption of multiple constellations by various international clients.

Of course, using data from different sensors has its challenges. At Edgybees, we seek to help our users overcome these challenges by allowing data from different constellations to have the same geospatial accuracy, formats, and meta data.  This is especially useful when leveraging AI as part of the analysis workflow.   This increase in satellite data will also facilitate more accurate predictive modeling.

3. Integration of Geospatial Data and technologies in ‘standard’ data workflows

In my time at Uber Maps, I saw geospatial data used in incredibly powerful ways combining traces from vehicles, municipality reports, and imagery together to form a cohesive picture. In 2025 we can expect geospatial data and workflows to expand its impact outside of the domain of the GIS office or mapping group. We will see geospatial data used interchangeably with structured non-geo data as part of the data science workflows increasingly relied on for making business decisions.

4. Rise of Geospatial Data Marketplaces

As the demand for geospatial data grows, we can expect to see a boom in geospatial data marketplaces by 2025. These platforms will allow organizations to buy, sell, and share geospatial datasets from various sources, including satellite imagery, sensor networks, and public records.

By 2025, these data marketplaces will be highly sophisticated, using AI to ensure data provenance, quality, and security. This will make it easier for businesses, governments, and researchers to access the data they need for their projects without having to develop the infrastructure to collect it themselves.

5. The Z-axis becomes commonplace

Traditional imagery analysis has often stayed in the 2D domain with updated imagery being the base currency. While there has always been some usage of stereo imaging, it has not been as commonplace as its collection was seen as expensive, using substantial satellite capacity.   3D technology plays a crucial role in geospatial analysis because it provides a more accurate and realistic representation of the Earth’s surface and its features. By incorporating elevation, depth, and volume, 3D models allow for a better understanding of spatial relationships, terrain, and infrastructure, which is essential for urban planning, environmental monitoring, military operations and disaster management.

There has been many advancements in the creation of advanced Digital Elevation Models (DEMs), Digital Surface Models (DSMs) and 3D models by companies such as Maxar, Airbus, Blackshark, and others. Further, the proliferation of high frequency constellations and UAS missions has lowered the effective cost of collecting current stereo imagery allowing the creation of these 3D data sets. Indeed, at Edgybees, we leverage updated Digital Elevation Models for object detection and geospatial image alignment to allow users to enjoy a truer and more accurate representation of the Earth.

In 2025, we can expect the 3D models to become more commonplace as data availability increases, compute becomes more accessible, and the tools for leveraging and visualizing 3D data become more powerful.

6 Bonus Generative AI has a big impact for 2026

A saying commonly attributed to Bill Gates says:

Most people overestimate what they can do in one year and underestimate what they can do in ten years.

We have seen amazing advances in generative AI in 2024 with Large Language Models (LLMs) becoming more commonplace in each of our lives with Gemini, ChatGPT, and Llama moving closer to our fingertips. While these tools have a high school understanding of geospatial data I expect them to have a broad impact on our industry in 2026 and 2027. Whether its new ways to visualize data, detect patterns, or draw conclusions, these tools have the potential to accelerate our workflows in the years to come but I wouldn’t expect it to come quite in 2025.

Conclusion

The geospatial sector in 2025 will be characterized by greater accessibility, higher accuracy, and deeper integration with other advanced technologies like AI. As the industry continues to evolve, geospatial data will become an even more integral part of decision-making in sectors like transportation, agriculture, urban planning, disaster management, and environmental monitoring.

About Author:

Dr. Shay Har-Noy (Shy) is a satellite data expert with leadership experience at top satellite and tech companies including Uber, ViaSat, Spire, and Maxar.  Shay is currently the Chief Executive Officer of Edgybees, a cutting edge company making aerial video and satellite imagery more accurate through ML/AI.  At DigitalGlobe/Maxar, he was responsible for their Platform business — their high growth effort to get DigitalGlobe’s 15 year digital library in the cloud and available for ML/AI. Prior to joining DigitalGlobe, Shay was founder and CEO of Tomnod combining crowdsourcing and machine learning to create new applications for satellite imagery. Tomnod was acquired by DigitalGlobe in early 2013. He also saw first hand the importance of accuracy in large scale automation through his experience at Uber leading the product group responsible for detecting and fixing map errors in real time.

Dr. Har-Noy graduated summa cum laude from Rice University and received a Ph.D in Electrical Engineering with research in image processing from UC-San Diego.

Tags: 3D geospatial models, AI in geospatial, geospatial data marketplaces, geospatial technology, geospatial trends 2025, geospatial workflows integration, high-resolution imagery, predictive modeling, satellite constellations, spatial analysis tools

Category: Industry Predictions

This entry was posted on Monday, January 13th, 2025 at 5:11 pm. You can follow any responses to this entry through the RSS 2.0 feed. You can skip to the end and leave a response. Pinging is currently not allowed.