Phillipe Mateu, our Geo Int expert at Prométhée Earth Intelligence, reveals everything there is to know about geospatial data.
What is geospatial data?
Prométhée Earth Intelligence uses earth observation satellites, the geospatial data is the pixel of an image and within this pixel we can have a lot of information that is brought in notably by the radiometry, the spectral refraction of this pixel. The assembly of this pixel can provide further information for detecting and recognizing elements. Like a boat, a car, a field. This concerns the image part, but we can also have geospatial data, which comes from electromagnetic sensors, i.e. all communication signals, radar and so on.
Any element in the world can be located with its coordinates, and in addition to this geospatial data, we add a notion of temporality, enabling us to track them over the entire globe, and thus know what’s happening at any given moment. In other words, geospatial data is a spatial and temporal reference frame.
What is a Geo Int expert?
Geo Int is the ability to analyze this data, this information, to merge it together, to turn raw data into intelligible, strategic information.
In short, Geo Int is a strategic decision-making tool.
As a Geo Int expert, what is your role at Prométhée Earth Intelligence ?
I have several roles, the first of which is to transform the capabilities of the Prométhée Earth Intelligence systems, including the satellites, the ground segment and the digital platform, into services that meet needs. My role is to adapt our systems so that our customers can make the best use of our facilities and capabilities.
My other job is to help optimize the tools and functionalities for our digital platform, EOP. Our aim is to democratize access to geospatial data, so it’s very important that our platform is accessible to everyone, without needing to be a Geo Int expert to use it.
There are several types of satellite. Is the geospatial data processing process the same for all types of satellite?
There are different types of satellite. There are those with active sensors and others with passive sensors. We use passive optical sensors, because they need the sun’s rays to operate.
And what about our future satellites?
We work in the visible spectrum (380 to 780 nm). Our first ProtoMéthée satellite, due to leave in October 2023, will use multispectral technology. Our second satellite, HyperMéthée, will carry a hyperspectral camera to complement ProtoMéthée’s vision.
The multispectral camera works in 7 spectral bands: from red to blue in the visible range, restoring the true color of images. The hyperspectral camera, on the other hand, analyzes several fine spectral bands to obtain a light spectrum and thus characterize the matter of objects on the ground.
How will the geospatial data from our first ProtoMéthée satellite be used?
There are two types of analysis: the so-called logical analysis, which is performed by the human eye. It’s a very reliable method, because the human eye, with experience, will recognize and work on certain objects fairly quickly. However, given the exponential increase in the number of images to be processed and the surface areas to be covered, we need to be able to enlist the help of what we call mathematical analysis.
It consists of automated image processing, ranging from algorithms that detect boats on an image, for example, to others that characterize the type of vegetation, the type of building, the condition of roads… These are all the elements that can be extracted from an image, more or less automatically. An image never lies, but you can make it say anything if it’s not properly analyzed!