Ship Detection

Ships at sea Detection detects ships in SPOT images from the SPOT 6/7 Display Block.

Technical information

This algorithm is based on machine learning. It is a blended version from Airbus-specific research and the results from a Kaggle competition launched in 2018. The algorithm runs on 768px by 768px tiles.

Supported Input

This block can be used with:

• SPOT Display (Streaming)-> Raster Tiling
• SPOT Display (Download) -> DIMAP -> GeoTIFF Conversion -> Raster CRS Conversion (CRS: EPSG:3857) -> Raster Tiling

SPOT Reflectance (Download) with Pan-Sharpening (UP42) is currently not supported.

Feature	Info
Input parameters	Tiled imagery in GeoTIFF with a size of 768 pixels by 768 pixels. Expected ground resolution at 1.20 m. (i.e. WebMercator projection). CRS: EPSG:3857
Output format	GeoJSON with each detected ship as a specific polygon.
Training dataset	The algorithm has been tested on various images including difficult images (i.e. with clouds and waves).

Performance

It is considered to work very well for medium & large ships (>26m) at sea. For smaller ships, the results will depend significantly from the quality of the image and the complexity of the scene.

Use cases

Shipping traffic is growing fast. More ships increase the chances of infractions at sea like environmentally devastating ship accidents, piracy, illegal fishing, drug trafficking, and illegal cargo movement. This has compelled many organizations, from environmental protection agencies to insurance companies and national government authorities, to have a closer watch over the open seas. This algorithm associated with satellite images enables precise and fast monitoring.

More information

For more information about this processing block, please see the provider website.