Deformation trend change detection for infrastructure monitoring

The Copernicus Sentinel-1 Synthetic Aperture Radar (SAR) mission has turned out to be a game-changer for the Earth Observation community. This constellation of two twin platforms is providing wide-scale, systematic (every 6 days) and access-free imagery over most of the globe.

SAR-derived information is used to produce millimetric-precision ground surface deformation maps. Thanks to Sentinel-1 SAR revisit time, new deformation maps can be delivered to end-users on a regular basis, showing average deformation rates of Persistent Scatter points and their displacement time series. Each information layer is made of hundreds of thousands of measurement points, and can be used for detecting significant instabilities on critical infrastructures, thus contributing to plan and optimize mitigation actions. Big Sentinel-1 data availability combined with cloud-based solution allows TRE Altamira to deliver even nationwide InSAR databases, offering millions of measurement points with a time series of displacement at every new update.

How can we help end-users to get rid of such massive information?

How can we contribute to the creation of a novel service that meets the monitoring need of a specific market sector?

Running the MATTCH project (ESA Open Call for Science, 2019), TRE Altamira has already implanted DL-based methodologies to big Sentinel-1 data and the detection of time series trend change for data screening purposes. So far, hotspots can be identified, but no indication about a possible reason and driving mechanisms for trend change is given to end-users. Moreover, how to provide a streamline of information that could be easily integrated with other sources of information in their workflow?

With this use case, we aim at creating a commercial service to monitor critical infrastructure as a sample. We will develop new Deep Learning architectures on trend change detection from dense InSAR point time series combined with industrial geodetic (GNSS) and other (e.g. drone) measurements to identify clusters of points sharing some attributes or features that are key for critical infrastructure monitoring at a large scale.

DeepCube aims to link any hotspots to a possible reason for trend change. Specifically, using Artificial Intelligence on InSAR data and sparse in-situ geodetic measurements for training and fusion, we aim at providing a label that indicates the reason for the change to each hotspot automatically detected and is impacting urban infrastructure, as well as critical infrastructure such as highways, tunnels, bridges, and airports.

Use Case Leader
Tre Altamira

Interested in learning more? Contact us!
Chiara Gervasi,