A New Dataset and Performance Benchmark for Real-time Spacecraft Segmentation in Onboard Computers

TL;DR

This paper introduces a comprehensive dataset of nearly 64,000 annotated spacecraft images with real-world challenges, and benchmarks YOLO models for real-time spacecraft segmentation on onboard computers, advancing autonomous space inspection.

Contribution

The authors created a large, diverse spacecraft image dataset with realistic noise and distortions, and provided performance benchmarks for YOLO-based segmentation models suitable for onboard space applications.

Findings

YOLOv8 and YOLOv11 achieved a Dice score of 0.92

Inference time was approximately 0.5 seconds

Dataset includes challenging real-world conditions

Abstract

Spacecraft deployed in outer space are routinely subjected to various forms of damage due to exposure to hazardous environments. In addition, there are significant risks to the subsequent process of in-space repairs through human extravehicular activity or robotic manipulation, incurring substantial operational costs. Recent developments in image segmentation could enable the development of reliable and cost-effective autonomous inspection systems. While these models often require large amounts of training data to achieve satisfactory results, publicly available annotated spacecraft segmentation data are very scarce. Here, we present a new dataset of nearly 64k annotated spacecraft images that was created using real spacecraft models, superimposed on a mixture of real and synthetic backgrounds generated using NASA's TTALOS pipeline. To mimic camera distortions and noise in real-world image acquisition, we also added different types of noise and distortion to the images. Our dataset includes images with several real-world challenges, including noise, camera distortions, glare, varying lighting conditions, varying field of view, partial spacecraft visibility, brightly-lit city backgrounds, densely patterned and confounding backgrounds, aurora borealis, and a wide variety of spacecraft geometries. Finally, we finetuned YOLOv8 and YOLOv11 models for spacecraft segmentation to generate performance benchmarks for the dataset under well-defined hardware and inference time constraints to mimic real-world image segmentation challenges for real-time onboard applications in space on NASA's inspector spacecraft. The resulting models, when tested under these constraints, achieved a Dice score of 0.92, Hausdorff distance of 0.69, and an inference time of about 0.5 second. The dataset and models for performance benchmark are available at https://github.com/RiceD2KLab/SWiM.