Lightweight Cloud Masking Models for On-Board Inference in Hyperspectral Imaging

TL;DR

This paper evaluates lightweight machine learning models, including CNNs with feature reduction, for cloud masking in hyperspectral satellite images, emphasizing high accuracy and efficiency suitable for on-board satellite processing.

Contribution

It introduces a lightweight CNN model with feature reduction that balances accuracy and computational efficiency for real-time hyperspectral cloud masking.

Findings

CNN with feature reduction achieved high accuracy (>93%)

Models with up to 597 parameters are effective for on-board deployment

Lightweight models enable real-time space-based hyperspectral image processing

Abstract

Cloud and cloud shadow masking is a crucial preprocessing step in hyperspectral satellite imaging, enabling the extraction of high-quality, analysis-ready data. This study evaluates various machine learning approaches, including gradient boosting methods such as XGBoost and LightGBM as well as convolutional neural networks (CNNs). All boosting and CNN models achieved accuracies exceeding 93%. Among the investigated models, the CNN with feature reduction emerged as the most efficient, offering a balance of high accuracy, low storage requirements, and rapid inference times on both CPUs and GPUs. Variations of this version, with only up to 597 trainable parameters, demonstrated the best trade-off in terms of deployment feasibility, accuracy, and computational efficiency. These results demonstrate the potential of lightweight artificial intelligence (AI) models for real-time hyperspectral image processing, supporting the development of on-board satellite AI systems for space-based applications.