Convolutional Neural Networks for Multispectral Image Cloud Masking

TL;DR

This paper evaluates various convolutional neural network architectures for cloud masking in multispectral remote sensing images, demonstrating their potential as effective alternatives to traditional feature-based methods.

Contribution

It introduces the application of CNNs to multispectral cloud masking and compares their performance with classical machine learning approaches.

Findings

CNN architectures outperform traditional methods in cloud masking accuracy.

Deep learning models require less manual feature engineering.

Results indicate CNNs are a promising tool for remote sensing cloud detection.

Abstract

Convolutional neural networks (CNN) have proven to be state of the art methods for many image classification tasks and their use is rapidly increasing in remote sensing problems. One of their major strengths is that, when enough data is available, CNN perform an end-to-end learning without the need of custom feature extraction methods. In this work, we study the use of different CNN architectures for cloud masking of Proba-V multispectral images. We compare such methods with the more classical machine learning approach based on feature extraction plus supervised classification. Experimental results suggest that CNN are a promising alternative for solving cloud masking problems.