The Descartes Labs Platform is the missing link in making satellite imagery useful. We collect data daily from public and commercial imagery providers, clean it, calibrate it, and store it in an easy-to-access file system, ready for scientific analysis.
Using our Python APIs, access any image of the Earth within seconds.
Input any area of interest
Find all matching imagery within our archive. Filter results by: • Start and end date • Satellite constellation • Satellite spectral band • Cloud cover
Create a composite
Generate a single composite image from your chosen range of imagery.
Retrieve a time stack of imagery
View how your area of interest changes over time. We have the full historical archive for Landsat 5/7/8, MODIS, Sentinel 1/2/3.
Calculate NDVI or any other custom statistic
In addition to providing the imagery, we’ll also run your calculations over each pixel within your area of interest.
The Descartes Labs Platform provides one-stop access to an unparalleled repository of pre-processed historical and real-time public and commercial satellite imagery.
Top of atmosphere reflectance
Currently, the Descartes Labs Platform ingests 5 terabytes (TB) of near real-time data per day, roughly equivalent to 5,000 hours of standard video. Our current corpus is over 5 petabytes of data (5,000 TB) with the ability to grow much larger. With sensor data growing exponentially, the Descartes Labs Platform is designed to respond elastically to this data explosion and harness it for real-time analysis.
Putting big data to work requires a platform that can ingest and process massive datasets. To tackle this problem, we built a supercomputer in the cloud.
The Descartes Labs Platform is built to ingest virtually any kind of data — not just satellite imagery — including weather information, commodity price histories, web crawls, and sentiment analysis from social media networks.
When conducting analysis or machine learning on these massive datasets, the Platform parallelizes large calculations automatically, scaling cloud resources on demand, in response to both calculation size and desired processing time. Thus, data collected over decades can be processed in hours.