AI detects ships onboard satellites to cut maritime data sent earthward
AI-Based On-Board Maritime Object Detection for Earth Observation Payload Data Reduction on Versal Embedded Hardware
Computer Vision and Pattern RecognitionArtificial Intelligence
Summary
Earth observation satellites collect huge amounts of data, but most of it does not contain ships, which are of interest for maritime surveillance. The authors created a new dataset and trained an AI model to spot vessels directly onboard the satellite. They deployed this AI on embedded hardware, so it can quickly process scenes and decide which parts contain ships worth sending to Earth. This approach cuts down data transmission significantly while keeping almost all ships visible in the images.
What this means in practice
- •For satellite operators: Reduce satellite downlink bandwidth by transmitting only maritime objects detected onboard, increasing effective data usage.
- •For embedded system engineers: Implement onboard AI vessel detectors on embedded DPUs to enable real-time maritime object detection with low power and latency.
Authors
Thomas Goudemant, Aurélien Bobey, Omar Hlimi, Marjorie Bellizzi
Abstract
Very-high-resolution Earth-observation satellites acquire more data than they can store and downlink, while in maritime surveillance the vessels cover a tiny fraction of each scene. We study onboard vessel detection as a way to select what is downlinked, which reduces the data according to its content rather than coding every pixel; it is complementary to conventional onboard compression. The work follows three axes. (i) Data and algorithm: a controlled dataset is generated from 68 annotated Maxar scenes with 43 vessel classes, and a YOLOX-S detector is trained on it. (ii) Embedded deployment: the detector is quantized and deployed on the DPU of a Versal VC1902, with a limited loss of detection quality and a processing time of a few seconds per scene. (iii) Data reduction: we propose several downlink modes, from metadata only (box, class and score of each detection) to image crops around vessels, tiles holding detections, or the whole scene with a degraded background, and estimate from the measured detection errors the trade-off each offers between the vessels kept and the volume downlinked. On our dense harbor and coastal scenes, tiles keep 98% of the vessels with 29% of the scene volume, and crops 83% with 3%.