A generic self-supervised learning (SSL) framework for representation learning from spectra-spatial feature of unlabeled remote sensing imagery

TL;DR

This paper introduces a novel self-supervised learning framework tailored for remote sensing imagery that leverages spectral and spatial features, significantly improving model performance on land cover and soil parameter tasks.

Contribution

The work presents a new SSL framework with two pretext tasks specifically designed for spectra-spatial data in remote sensing, addressing limitations of RGB-based pretext tasks.

Findings

Significant performance improvements on land cover classification.

Enhanced accuracy in soil parameter retrieval.

Effective learning from unlabeled spectra-spatial remote sensing data.

Abstract

Remote sensing data has been widely used for various Earth Observation (EO) missions such as land use and cover classification, weather forecasting, agricultural management, and environmental monitoring. Most existing remote sensing data-based models are based on supervised learning that requires large and representative human-labelled data for model training, which is costly and time-consuming. Recently, self-supervised learning (SSL) enables the models to learn a representation from orders of magnitude more unlabelled data. This representation has been proven to boost the performance of downstream tasks and has potential for remote sensing applications. The success of SSL is heavily dependent on a pre-designed pretext task, which introduces an inductive bias into the model from a large amount of unlabelled data. Since remote sensing imagery has rich spectral information beyond the standard RGB colour space, the pretext tasks established in computer vision based on RGB images may not be straightforward to be extended to the multi/hyperspectral domain. To address this challenge, this work has designed a novel SSL framework that is capable of learning representation from both spectra-spatial information of unlabelled data. The framework contains two novel pretext tasks for object-based and pixel-based remote sensing data analysis methods, respectively. Through two typical downstream tasks evaluation (a multi-label land cover classification task on Sentienl-2 multispectral datasets and a ground soil parameter retrieval task on hyperspectral datasets), the results demonstrate that the representation obtained through the proposed SSL achieved a significant improvement in model performance.