Capella’s Intuitive Self-Serve Online Platform
Note: this is the second post of a two-part blog series; in the first blog post I discussed Capella’s Advanced SAR Imagery Products that will be accessible to customers via Capella’s online platform.
In my previous blog post I discussed how Capella will provide the most frequent, timely and high-quality synthetic aperture radar (SAR) imagery products available in the market today. But the only way this SAR imagery will deliver value to customers is if the data products are easily accessible. Furthermore, in order to enable the high-cadence monitoring use cases that benefit from SAR’s 24-hour all-weather Earth observation imaging, customers must be able to obtain the imagery products in an affordable manner with flexible on-demand access.
Capella’s Online Platform
This is why Capella has built a modern, reliable and scalable online platform for customers to access our SAR imagery that was built from the ground up using a cloud-native architecture. Our online platform provides both a web-based portal (Capella Console) and corresponding application programming interface (Capella API). These platform components allow customers to immediately discover, order and download existing archival imagery products in a self-serve fashion without any manual order processing delays. Capella’s system is built with end-to-end automation from order to delivery with fully-automated data processing and quality assurance so that customers receive their SAR imagery products as fast as possible.
One of the primary goals for our online platform is to provide an intuitive and frictionless user experience so customers can access the SAR imagery products they need in an efficient manner. Both the Capella Console web application and API make it easy for customers to search our image library catalog, filter archive results, order existing imagery products, submit new acquisition tasking requests and manage their accounts.
Since our SAR imagery products are distributed as Cloud Optimized GeoTIFF (COG) format, which enables efficient tile-based access to small portions of the overall dataset, the Capella online platform is able to provide corresponding capabilities that benefit our customers. For example, the Capella Console web application provides a streaming full-resolution image preview so users can get a real sense of image quality and scene content before ordering (as opposed to a static low-resolution thumbnail).
The Capella API uses an interoperable REST style with a variety of flexible and responsive endpoints (catalog search, data access, etc.) that can be used to integrate our services with other online platforms, enterprise systems or government architectures. Furthermore, Capella’s compatible RESTful web services allow developers to automate complex workflows and orchestrate multi-source analytics. This enables users to automate production of derived information products or extract geospatial intelligence (GEOINT) that will help analysts discover and characterize relevant patterns in a timely manner. Finally, our mission operations and online platform are built with the highest levels of security and confidentiality so customers can trust that their interactions with our system are private and anonymized.
New Acquisition Tasking
You may notice the “Add Tasking Request” button in the previous screenshot above – at Capella Space we are putting the power directly into the hands of our customers to task our constellation! Both the Console web application and API allow users to submit a new acquisition tasking request which provides on-demand self-serve satellite commissioning. Capella’s tasking scheduler system is fully-automated which provides responsive feasibility analysis and definitive accepted/rejected response within 15 minutes or less. Once a new acquisition tasking request has been accepted, users can use the platform to check on the real-time status of their new imaging collections.
Once a new tasking request is accepted the new imagery will always be acquired within the collection window duration selected by the user. Furthermore, Capella delivers new imagery collections to customers as fast as possible via automated processing of raw data into the SAR imagery products delivered. In other words, with Capella’s fully-automated self-serve tasking system there is no sending an email to an order desk, waiting hours or days for a response, then receiving your imagery products via old-fashioned FTP or in the mail. Finally, in the future we will be adding functionality to the platform which will help customers discover potential new acquisition opportunity candidates for their location and time period in order to maximize the success of new collection campaigns.
Enabling our Developer Community with TileDB Cloud
In many situations users are not necessarily interested in working with the SAR imagery itself, instead their projects benefit more by having access to value-added services or the analytics-derived information that can be generated from SAR data. Consequently, Capella has created a SAR developer community program where we collaborate with a broad community of the brightest data scientists and software engineers. Many of the tools and algorithms developed by this community are open source and can be found on Capella’s GitHub repository.
We also want to make it as easy as possible for data scientists and software developers to access a generous collection of free and open sample datasets so they can build novel analytics and machine learning models that extract insights from SAR imagery. This is why Capella has contributed a collection of our airborne campaign datasets to the SpaceNet 6 challenge focused on multi-sensor all-weather mapping. Moreover, after we launch our first operational satellites we will release a generous collection of open data in order to foster the development of analytics that are tuned to Capella’s SAR imagery. However, when running analytics on SAR imagery it is not efficient to download all of the full datasets to a local computing environment for processing. Consequently, in order to provide efficient cloud-based access to our SAR imagery along with useful developer resources we are partnering with TileDB, Inc. on a project to make our open data available in their TileDB Cloud environment for members of our community.
The TileDB database is a powerful, open-source, cloud-native storage engine for accessing raster data as multi-dimensional arrays which is particularly well-suited to Capella’s high-cadence spatio-temporal stacks of SAR imagery. Furthermore, TileDB Cloud allows our developer community to access and process multi-dimensional data at scale with helpful productivity resources such as hosted Jupyter Notebooks to perform exploratory data analysis. TileDB Cloud is a perfect environment for data scientists who want quick access to Capella’s SAR imagery as analysis-ready data arrays that are optimized for scalable computations and parallel distributed processing.
TileDB provides developers API-based access to Capella’s open SAR data as numerical arrays that can be processed using programming languages such as R or Python. With a variety of data science toolkit integrations and hosted Jupyter Notebooks for interactive computing, TileDB Cloud allows the developers in our community to rapidly prototype and execute a wide range of different analysis methodologies from simple statistics to advanced machine learning. Furthermore, the Jupyter Notebooks environment makes it easy to create and share documents that contain live code, equations, visualizations, and narrative text that will help facilitate team collaboration within the Capella Space developer community.
In summary, Capella’s intuitive self-serve online platform will provide our customers with a wide variety of functionality which will increase user productivity, profitability and mission success. Whether it is a student searching our image library for a dataset over their study area, an analyst tasking our satellite constellation to acquire new imagery for a specific location & time, or a data scientist building a machine learning model to extract features from our SAR data, the Capella platform provides the functionality to make them successful. Finally, in addition to hosting our SAR imagery products, our online platform can also be used to aggregate and disseminate a variety of geospatial content derived from Capella data such as automated change monitoring alerts, object detections, surface deformation maps, etc. which will unlock the most value from our SAR data.
If you are interested in learning more about Capella’s online platform or get access to the open SAR data available in TileDB Cloud please join our SAR developer community.