Single Image Cloud Detection via Multi-Image Fusion

TL;DR

This paper introduces a novel method for detecting clouds in single images by leveraging multi-image fusion techniques, reducing the need for annotated data and enhancing detection accuracy.

Contribution

It proposes a new approach that uses multi-image fusion to bootstrap single image cloud detection, demonstrating improved performance with less annotated data.

Findings

Reduces reliance on annotated training data.

Improves cloud detection accuracy.

Leverages image quality estimation for cloud detection.

Abstract

Artifacts in imagery captured by remote sensing, such as clouds, snow, and shadows, present challenges for various tasks, including semantic segmentation and object detection. A primary challenge in developing algorithms for identifying such artifacts is the cost of collecting annotated training data. In this work, we explore how recent advances in multi-image fusion can be leveraged to bootstrap single image cloud detection. We demonstrate that a network optimized to estimate image quality also implicitly learns to detect clouds. To support the training and evaluation of our approach, we collect a large dataset of Sentinel-2 images along with a per-pixel semantic labelling for land cover. Through various experiments, we demonstrate that our method reduces the need for annotated training data and improves cloud detection performance.