Transfer Learning for Onboard Cloud Segmentation in Thermal Earth Observation: From Landsat to a CubeSat Constellation

Niklas Wölki, Lukas Kondmann, Christian Mollière, Martin Langer, Julia Gottfriedsen, Martin Werner

Keywords: Transfer Learning, Cloud Detection, Domain Adaptation, Computer Vision, CubeSat, Onboard processing

TL;DR: We improve thermal-only onboard cloud segmentation on a resource-constrained CubeSat by jointly training on public Landsat data and mission-specific samples, enabling accurate, real-time onboard inference with a lightweight UNet.

Abstract: Onboard cloud segmentation is a critical yet underexplored task in thermal Earth observation (EO), particularly for CubeSat missions constrained by limited hardware and spectral information. CubeSats often rely on a single thermal band and lack sufficient labeled data, making conventional cloud masking techniques infeasible. This work addresses these challenges by applying transfer learning to thermal cloud segmentation for the FOREST-2 CubeSat, using a UNet with a lightweight MobileNet encoder. We pretrain the model on the public Landsat-7 Cloud Cover Assessment Dataset and fine-tune it with a small set of mission-specific samples in a joint-training setup, improving the macro F1 from 0.850 to 0.877 over FOREST-2-only baselines. We convert the model to a TensorRT engine and demonstrate full-image inference in under 5 seconds on an NVIDIA Jetson Nano. These results show that leveraging public datasets and lightweight architectures can enable accurate, efficient thermal-only cloud masking on-orbit, supporting real-time decision-making in data-limited EO missions.

Submission Number: 13