The algorithm uses object detection in Pléiades display imagery to detect and classify multiple types of aircraft. The result is a GeoJSON file with vertically oriented polygons drawn around detected aircraft.
This process detects and classifies multiple types of aircraft:
- Carrier aircraft
- Fighter aircraft
- Helicopters
- Small aircraft
- Other
Aircraft detection can be used for aircraft management, airport monitoring, and environmental modeling.
An input Pléiades image of Heathrow Airport (UK)
Detected aircraft polygons drawn on the input image
The algorithm has been developed using 2,500 satellite images across 350 air bases in 50 countries spanning 6 continents. The algorithm counted 54,000 aircraft to create a historical baseline of activity and detect ongoing changes.
For optimal results, the algorithm must be applied to imagery captured over airport areas. You can apply it to similar areas, but the results might not be as accurate.
Scenarios represented in the training data:
- Different times of day
- Different times of year
- Different terrains
- Different object configurations
The data item must be a supported data product:
- Pléiades catalog: Display
The data item must be CNAM-compatible. Check that the data item has been added to storage in 2023 or later.
You must specify the data items you want to apply the process to.
You must specify the title of the output data item.
Use the detection-aircraft-oi name ID for the processing API.
{ "inputs": { "title": "Processing imagery over Berlin", "item": "https://api.up42.com/v2/assets/stac/collections/21c0b14e-3434-4675-98d1-f225507ded99/items/23e4567-e89b-12d3-a456-426614174000" }}| Parameter | Overview |
|---|---|
inputs.title | object | required The title of the output data item. |
inputs.item | object | required A link to the data item in the following format: |