Embedding Earth: Self-supervised contrastive pre-training for dense land cover classification

by   Michail Tarasiou, et al.

In training machine learning models for land cover semantic segmentation there is a stark contrast between the availability of satellite imagery to be used as inputs and ground truth data to enable supervised learning. While thousands of new satellite images become freely available on a daily basis, getting ground truth data is still very challenging, time consuming and costly. In this paper we present Embedding Earth a self-supervised contrastive pre-training method for leveraging the large availability of satellite imagery to improve performance on downstream dense land cover classification tasks. Performing an extensive experimental evaluation spanning four countries and two continents we use models pre-trained with our proposed method as initialization points for supervised land cover semantic segmentation and observe significant improvements up to 25 initialization, especially so when ground truth data are scarse. Through a series of ablation studies we explore the qualities of the proposed approach and find that learnt features can generalize between disparate regions opening up the possibility of using the proposed pre-training scheme as a replacement to random initialization for Earth observation tasks. Code will be uploaded soon at https://github.com/michaeltrs/DeepSatModels.


page 1

page 4

page 6

page 9

page 15

page 16


SSL4EO-S12: A Large-Scale Multi-Modal, Multi-Temporal Dataset for Self-Supervised Learning in Earth Observation

Self-supervised pre-training bears potential to generate expressive repr...

Ben-ge: Extending BigEarthNet with Geographical and Environmental Data

Deep learning methods have proven to be a powerful tool in the analysis ...

DeepSatData: Building large scale datasets of satellite images for training machine learning models

This report presents design considerations for automatically generating ...

Fair contrastive pre-training for geographic images

Contrastive representation learning is widely employed in visual recogni...

DynamicEarthNet: Daily Multi-Spectral Satellite Dataset for Semantic Change Segmentation

Earth observation is a fundamental tool for monitoring the evolution of ...

Spotting Virus from Satellites: Modeling the Circulation of West Nile Virus Through Graph Neural Networks

The occurrence of West Nile Virus (WNV) represents one of the most commo...

Please sign up or login with your details

Forgot password? Click here to reset