top of page

FarEarth Edge

​

FarEarth Edge pre-processes raw data to AI-ready imagery onboard the satellite. High-quality input imagery makes your real-time low latency applications possible.

​

FarEarth Edge seamlessly integrates with the rest of the FarEarth platform. Downlinked data is processed by FarEarth Cloud on-ground to continuously calibrate the satellite imager. These models are periodically uplinked to the satellite for accurate onboard pre-processing.

 

FarEarth Edge pre-processing produces radiometrically calibrated, band-aligned, terrain-corrected and precision-geolocated imagery. This is important to provide high-quality input for your edge processing application.

 

While edge processing can minimise data downlink volumes, users often still require high-quality, map-ready imagery of detected events for further analysis. Relevant raw data is downlinked and precision-processed on the ground to give context and inform decision-making. These images can be used for AI training and model refinement.

Edge

Edge processing is a trade-off between the accuracy and quality of the image, and the latency and cost of processing onboard. Where this balance lies depends on your specific application.

FarEarth Edge

FarEarth Cloud

image.png

Systematic Edge processing

Image processing with calibrated physical model

  • Temperature compensated radiometric correction

  • Top-of-atmosphere reflectance

  • Lens distortion correction

  • Noise filtering

  • Precision band alignment

  • Calibrated geolocation using a sensor model and satellite attitude and ephemeris data

image.png

Precision Edge processing

Precision refinement using reference imagery

  • Processing with calibrated physical model

  • Terrain correction using a DEM

  • Precision pixel alignment with references

image.png
image.png

Best quality image

  • Very high accuracy

  • High-throughput parallel processing

  • Full-featured platform with data management

bottom of page