Introduction
Over the last year, SpaceKnow had the privilege to enhance its commercial offering with support from the ESA InCubed program. The GEMSTONE (Global Economy Monitoring System delivering Transparency and Online Expertise) project. Complementing SpaceKnow’s Machine Learning based economic activity indicators, derived from satellite data. With GEMSTONE, SpaceKnow aims to expand its capabilities.
SpaceKnow capabilities
Internally developed supervised and unsupervised algorithms provide detections on satellite imagery over monitored areas of interest in an automated and scalable manner. Using various data fusion and time series modeling techniques these detections are then aggregated into economic indices. In GEMSTONE, SpaceKnow developed eight new algorithms to detect the following raw materials and man-made structures: Coal, Iron, Lithium, Wood, Open-Pit Mines, Containers, Oil Tanks, and Roads. The algorithms can be used to sweep over large areas to find those 8 objects. SpaceKnow can then monitor changes in those objects to provide a time series of developments.
The indices are delivered to the customer through API or interactive dashboards, allowing clients to instantly and without any bias see the performance of a specific industry in a selected region (for instance Coal Mining in China).
Use Case: Nagoya
Nagoya is the largest port in Japan, accounting for 10% of total trade value. Several supplies are stored outdoors at Nagoya, making it a great place to showcase the power of the algorithms. The focus is on the land-use segmentation and GEMSTONE algorithms for this example.
In Figure 1 below, analysis is presented for the port and city of Nagoya. The timeline at the bottom displays quantitative changes in each of the segmented elements. The outputs of the SpaceKnow segmentation algorithms can be observed: orange for Urban, green for Non-Urban, purple for Roads, and Blue for Water.
Zooming into Nagoya port in Figure 2, the detailed view on the left displays oil tank detections. Single oil tanks are highlighted in red; blocks of smaller oil tanks are highlighted in blue. The sub-polygons drawn around the oil tanks give precise insight into small sections of Nagoya port; separate time series showing the development of oil tank counts for the sub-polygons can be generated. On the right-hand side, detections of wood (orange) and containers (pink) are shown.
Key Takeaways:
- In cooperation with ESA, SpaceKnow is expanding its automated, scalable solution for near-real-time monitoring of economies, companies, and supply chains.
- Customers can access the insights through an online platform, without any initial investment or training needed.
Interested in learning more? Reach out to our team at info@spaceknow.com
