SSL4EO-L: Datasets and Foundation Models for Landsat Imagery

TL;DR

This paper introduces SSL4EO-L, a large self-supervised learning dataset for Landsat imagery, along with foundation models and benchmarks, to advance deep learning applications in Earth observation.

Contribution

It presents the first self-supervised learning dataset and foundation models for Landsat imagery, enabling improved remote sensing analysis and reproducibility.

Findings

Pre-trained models show strong performance on semantic segmentation tasks.

Largest Landsat dataset with 5 million image patches for self-supervised learning.

Benchmark datasets facilitate future research and comparison.

Abstract

The Landsat program is the longest-running Earth observation program in history, with 50+ years of data acquisition by 8 satellites. The multispectral imagery captured by sensors onboard these satellites is critical for a wide range of scientific fields. Despite the increasing popularity of deep learning and remote sensing, the majority of researchers still use decision trees and random forests for Landsat image analysis due to the prevalence of small labeled datasets and lack of foundation models. In this paper, we introduce SSL4EO-L, the first ever dataset designed for Self-Supervised Learning for Earth Observation for the Landsat family of satellites (including 3 sensors and 2 product levels) and the largest Landsat dataset in history (5M image patches). Additionally, we modernize and re-release the L7 Irish and L8 Biome cloud detection datasets, and introduce the first ML benchmark datasets for Landsats 4-5 TM and Landsat 7 ETM+ SR. Finally, we pre-train the first foundation models for Landsat imagery using SSL4EO-L and evaluate their performance on multiple semantic segmentation tasks. All datasets and model weights are available via the TorchGeo (https://github.com/microsoft/torchgeo) library, making reproducibility and experimentation easy, and enabling scientific advancements in the burgeoning field of remote sensing for a multitude of downstream applications.