Today, the Maxar constellation collects 3 million sq km of new high-resolution imagery every single day, adding to the depth and breadth of our 20-year, 110+ petabyte imagery archive. When WorldView Legion begins operations, the amount of daily imagery collected will increase to 5 million sq km with some areas imaged up to 15 times a day. That is a lot of data to organize and prepare for analysis! It is also why the concept of “analysis-ready data” resonates so strongly with us—both as providers and users of our industry-leading imagery.
With the new Maxar Analysis-Ready Data (ARD) product, we’ll organize and process the imagery so that you can jump straight into extracting features, detecting objects or unlocking the answers to your unique use case with our imagery. But what exactly does "analysis-ready" mean?
At Maxar, we identified five features as key requirements for satellite imagery to be considered analysis-ready; they serve as the foundation of our Maxar ARD product:
- Atmospheric and radiometric correction
- Orthorectification and alignment
- Optimized for cloud computing
- Configured for localized analysis—metadata and data masks
- Accelerated access and direct delivery
Atmospheric and radiometric correction
My colleagues have previously explained the importance of radiometric calibration and Maxar’s patented Atmospheric Compensation technology for satellite imagery analytics, so it should come as no surprise to see these corrections listed as a key requirement for analysis-ready imagery. Without compensating for atmospheric effects or calibrating the spectral bands, it is more difficult to correctly distinguish between relevant and nonrelevant features in the imagery or to rapidly analyze images across time and space.
With multiple processes available across the industry for handling these corrections, it is easy to inadvertently analyze a time-series stack of images with wildly varying colors and consistency, which leads to incorrect results and the need to develop location-specific artificial intelligence and machine learning (AI/ML) workflows. Maxar ARD eliminates this problem by providing consistent inputs ready for you to analyze immediately upon delivery.
We also apply our Dynamic Range Adjustment process to the pansharpened RGB imagery delivered with Maxar ARD to help further color-balance the imagery and enable rapid identification of objects with unique color signatures or those in more challenging locales, such as under trees or along the edge of an image.
Orthorectification and alignment
Similar to atmospheric and radiometric correction, orthorectification and alignment are needed to understand which features in the imagery are relevant. Orthorectification ties images to the correct location on Earth, and alignment controls for shifts between roads, buildings and other features within a stack of images taken from different angles. Orthorectification and alignment provide higher accuracy when detecting change, extracting features or monitoring transient objects, such as cars, ships and planes.
These processes reduce false negatives and false positives from automated exploitation workflows. Misaligned or poorly orthorectified imagery can cause you to detect two buildings where only one exists or instruct a driver to make an off-road turn onto the sidewalk.
Maxar ARD handles orthorectification for you, and it offers our proprietary Bundle Block Adjustment (BBA) process to further refine the alignment of images within the ordered stack. BBA can be deployed on up to 50 images to create a deep temporal stack of aligned imagery. The deeper stack combines signals to provide further confidence in automated object detections and allows you to more accurately analyze changes over time.
Optimized for cloud computing
To be truly “analysis-ready,” data must be delivered in a quickly readable format. Increasingly, this format must be optimized for cloud computing as AI/ML becomes a cornerstone of your work.
To enable faster computer processing, analysis-ready imagery should be delivered as cloud-optimized GeoTIFFs (COGs) and the accompanying metadata should be organized as a SpatioTemporal Asset Catalog (STAC). These are standard formats rapidly being adopted by the community.
COGs organize pixels into tiles, enabling key subportions of the image to be easily retrieved through HTTP GET range requests. Unlike with traditional GeoTIFFs, the subportion of the image does not need to be cached. This helps increase the efficiency of cloud computing environments to help lower your cloud computing costs. COGs are also agnostic towards compute environments, allowing customers to choose the best fit for their work.
Although version 1.0 of the STAC specification was released in January, we believe the common language promised by STAC will enable better data interoperability and speed up computing workflows. With Maxar ARD, your delivered order includes imagery in COGs format and STAC metadata.
Configured for localized analysis—metadata and data masks
Typically, satellite imagery is collected over large areas as a strip. Metadata, including cloud cover, is calculated for the entire strip, which makes it very easy to dismiss an image as not fit for analysis when looking for imagery over consistently cloudy areas or over a very small area of interest (AOI). Maxar ARD leverages your submitted AOI to configure the metadata and assist you in deciding which images should be considered fit for your custom analytic workflow.
The localized metadata is used to generate data masks, including one for clouds. This enables you to control for atmospheric conditions within an image that may affect your analysis. These data masks along with the imagery are clipped to your AOI—providing only what is needed to fulfill your mission.
Accelerated access and direct delivery
Imagery can be perfectly aligned, calibrated and optimized for a cloud-based workflow, but all those benefits are lost when you are forced to use cumbersome ordering and delivery processes. As with the requirement to optimize for cloud computing, data must be quickly discoverable and orderable to be “analysis-ready.”
APIs can enable faster analysis of satellite imagery through automation, batch ordering and machine-to-machine communication. As mentioned in our previous blog post about ARD, Maxar ARD includes an API for on-demand searches and orders of imagery in the Maxar ARD format. Direct delivery into your compute environment speeds up your analysis because you don’t have to wait for a download website to be stood up or switch between systems to pull in your order. Increased access and productivity can also be achieved by leveraging the software development kit included with Maxar ARD, which enables seamless integration with your workflow.
With these five features, Maxar ARD turns satellite imagery into analysis-ready data.
Want to learn more about Maxar ARD? Visit the Maxar ARD product page to see examples of Maxar ARD in action and to download a sample dataset.