Recent breakthroughs in mathematics make the impossible possible

Imagine being able to take a 3D digital stereoscopic image of anything on the Earth’s surface from orbit, and then rendering a model that could be used to cover anything from electrical powerlines to the Notre Dame fire. This vision is now reality. Kayrros is unleashing a new innovation in 3D mapping that combines high-resolution image processing and a whole lot of cutting-edge math to create a digital twin of any piece of the world’s infrastructure from a constellation of orbiting sun-synchronous satellites. After acquiring two satellite images pointed at the same spot on Earth, Kayrros applies a unique computational algorithm to transform those images into a 3D digital model, generating a digital topographical snapshot in near realtime. This unrivaled technology, made possible by advanced innovations in mathematics, holds the potential to monitor virtually any resource, structure, or asset as a 3D model, from stockpiles to emergency disasters, and agricultural development to digital models of the Empire State Building.

Rome wasn’t built in one day. Neither was 3D mapping from space. Years ago, Kayrros in-house satellite experts had the idea of reconstructing the Earth from the sky, and are now the global leaders in the field with expertise such as a PhD on the subject from France’s Ecole Normale Supérieure and 1st place in a US intelligence competition on 3D reconstruction. Backed by a team of global experts at France’s CMLA, France’s leading Research Center for Applied Maths, the project was driven by innovation and propelled by the necessary resources to push it forward. Its roots are grounded in advanced mathematics and computation algorithms, which form the foundations of Kayrros itself.

3D Mapping

2D Rendering of a 3D shipyard (Sources: Kayrros, CMLA)

Though satellite images are taken thousands of miles away from the Earth and processed through complex computational algorithms, the basic concept of 3D reconstruction connects to a much closer lens — the human eye, best illustrated through a concept that many might recall from teenage-era science class: parallax. Combining two slightly different optical viewpoints of the same object allows us to perceive a level of depth. This is what our brains do, thanks to the images of our two eyes. If an object is farther away from the lens, it will appear to move less; when closer, it will appear to move more. Computer vision scientists have turned this concept into algorithms, with multiple cameras acting as ‘multiple eyes.’ Kayrros in-house data scientists calculate the exact distances between the satellite lens and surface of the Earth to construct topographical models of the surface, calculating varying levels of depth with enough precision to draw a picture.

Kayrros sources its images from a unique constellation of optical satellites operated by Planet, the leading commercial satellite company that images the Earth every day to bring actionable global change, that fly in several sun-synchronous orbits around the earth in about 100 minutes. In the case of 3D reconstruction, two factors are particularly important: timing and angle. The agility of these satellites allows them to quickly turn their cameras up to 30 degrees away from the vertical, allowing for two images to be taken of the same object from two different angles — in a span of 10–30 seconds. This generates two pictures from the same place, with one seen from the left and the other from the right.

The innovation presents a new way of seeing what’s all around us on the Earth’s surface

The innovation presents a new way of seeing what’s all around us on the Earth’s surface. Digital twins of entire cities, complex infrastructures, and integrated engineering systems can be rendered in a span of hours. This is strengthened by newfound access to high-resolution commercial satellite imagery; previous developments in 3D mapping have been drone-focused. Where drones can generate complications — namely the need for human operators, a potentially slowing factor — satellite imagery steps in to fill the gaps.

The new generation of 3D mapping not only has the capacity to monitor events as they happen, but can track assets in a time series and may even prevent disastrous events through an algorithmic alert system. Kayrros bespoke mapping has the capacity to provide automatic monitoring of large areas with machine-learning, using low-resolution images to identify risk areas and generating notifications on change, eliminating reliance on reports for asset volumes that are often lagged or inaccurate — for companies themselves and other assets across the global industrial board.

The technology is also applicable to realtime events, such as natural disaster response. Should a city be hit by a natural disaster, digital modeling flags collapsed infrastructure to give emergency response teams a birds-eye-view of the ground to spot evacuation routes and catch pockets where people are in distress. The modeling can then be used to evaluate damage on the environment over time, with the original model serving as a detailed reference for pre-damaged topography and infrastructure.

3D Mapping will change the scope of Earth Observation, providing a new vantage point from which to digitally monitor what’s around us. And this is just the start.