o transform people’s understanding and decision-making about the planet, the Google Maps Platform is increasingly leveraging innovation from the advanced geospatial artificial intelligence in Google Earth AI to expand its Imagery Insights portfolio. The firm has now announced the launch of an experimental version of Earth AI-powered Aerial and Satellite Insights based on high-resolution 3D imagery.
To complement Street View Insights, the Imagery Insights portfolio on the Google Maps Platform is being expanded with the launch of Aerial and Satellite Insights (Experimental). According to Google, Aerial and Satellite Insights will put global, high-resolution imagery at people’s fingertips and broaden analysis from the fine details of a city street to a bird’s-eye view of the planet. Through the high resolution and angular view diversity of 3D aerial imagery or the flexible freshness of high-resolution satellite imagery, users will be empowered to make data-driven decisions at any scale covering a wide range of tasks.
Google expects that Aerial and Satellite Insights will empower organizations to revolutionize site selection and asset management. It enables more comprehensive site analysis and automated inventory audits, using historical data to detect changes and predict patterns over time. It can also support a range of applications and objectives. For example, when combined with digital terrain models, the approach enables telecommunication professionals to optimize wireless network planning and minimize signal obstructions. In another example, thanks to the integration of high-resolution, building-level data on solar potential and existing solar arrays, part of Earth AI, utilities and service providers will be able to recognize untapped market opportunities, accelerate renewable energy adoption and optimize network planning.
Aerial and Satellite Insights is integrated with Gemini Enterprise Agent Platform, enabling operational scaling while reducing in-the-field presence, risk and costs. Moreover, by utilizing BigQuery and Earth Engine, users can combine their proprietary data with high-resolution imagery context for detailed spatial analysis.
To expand the Google Maps catalogue of aerial and satellite imagery and bring Aerial and Satellite Insights to organizations at scale, the firm has partnered with Airbus and Vexcel. “Our expanded partnership with Google underscores Airbus’ commitment to delivering the highest quality and freshest data to our partners. By adding Airbus’ industry-leading satellite imagery to Google’s Aerial and Satellite Insights, we enable enterprise customers to move beyond visualization toward true decision-making with deeper, data-driven insights about the world in near real time,” comments Eric Eve, head of space digital at Airbus Defense and Space.
“Vexcel has built an extensive aerial imagery and geospatial data library spanning 45+ countries, which serves as a foundational layer for real-world AI. In close collaboration with Google Maps Platform, we’re discovering new ways to semantically understand everything that can be seen in the physical world, analyse it and generate knowledge at scale. Together, we’re unlocking new imagery-derived insights that were difficult, costly or even impossible to access before. Google brings world-class expertise in geospatial technology and AI that is also opening the door to a broader set of non-technical, non-geospatial users – and we see even more possibilities ahead,” says Erik Jorgensen, chairman and CEO of Vexcel.
Moreover, Google is introducing Aerial and Satellite Models (Experimental) within Google Cloud’s Model Garden. Developed as part of Google Research’s Remote Sensing Foundation effort, these models are designed to transform geospatial workflows by enabling users to extract planetary-scale insights without the need to train or maintain complex models themselves. Once again, besides utilizing the online catalogue, users can also bring their proprietary aerial or satellite imagery to the models and apply Google’s advanced AI to unlock insights from the data they already own.
Thanks to the advanced zero-shot and open vocabulary analysis capabilities, users can identify objects by simply describing them (e.g. ‘find all wind turbines’) and perform semantic retrieval to locate complex scenes (such as ‘a river running through a city centre’) using natural language. Additionally, the models can enable powerful semantic change detection to clarify the meaning and direction of changes. Examples include distinguishing between a long-standing and newer ‘industrial factory’, or identifying project phases like ‘pre-construction’ and ‘active construction’. This multi-modal capability contributes to a better understanding of the broader semantic and temporal context of the Earth rather than just its individual components, enabling more proactive and data-driven decisions across a wide range of critical use cases.
One of the many applications of Aerial and Satellite Insights is helping telecommunication professionals optimize wireless network planning by identifying terrain, buildings and other potential sources of signal obstruction. (Image courtesy: Google)
https://www.gim-international.com/content/news/google-maps-platform-adds-earth-ai-powered-aerial-and-satellite-insights
