Descarteslabs.com Website Review

Machine learning and remote sensing science show what pixels contain clouds The primary direction of a slope, which can affect temperature, moisture, and growth Cloud penetrating radar can observe the planet in any weather and at night Cloud-free Landsat 8 A visible spectrum composite image shows the earth as we see it Instant access to science-ready imagery and intelligence from multiple data sources. For decades, sensors have been collecting data about our planet. Yet complex global systems like agriculture, deforestation, water cycles, and many others, affect billions but are still poorly understood. Not only are these systems critical to understand for the good of humanity, they're also important for businesses as they adapt to — and attempt to thrive in — an ever-changing environment. We are seeing an exponential rise in the number of real-world sensors, producing ever more data. Combining multiple sources provides a far richer view than that of any single data source. To harness this power and enable global-scale computation, it's necessary to have a data refinery that combines data from diverse sources, cleans it up and makes it ready for modeling - and a platform to upon which to build living, learning models. That's exactly what we've built at Descartes Labs. Using synthetic aperture radar (SAR), we developed a proprietary model that can identify new construction starts on the ground on a monthly basis, regardless of weather conditions. This model enables a real-time look at changes and trends impacting infrastructure growth. Leveraging our database of industry leading high-resolution imagery, we built a model that first identifies field boundaries and then clas*ifies which crops are growing within each field. With this optimized approach, field teams spend less time surveying ground data Using high-resolution Airbus imagery, we built a computer vision model that can quickly identify all physical wind turbine as*ets worldwide in just a few hours. This solution automates an*lysis that would take a fleet of human an*lysts several months to complete.