Deep Unsupervised Embedding for Remotely Sensed Images Based on Spatially Augmented Momentum Contrast

Convolutional neural networks (CNNs) have achieved great success when characterizing remote sensing (RS) images. However, the lack of sufficient annotated data (together with the high complexity of the RS image domain) often makes supervised and transfer learning schemes limited from an operational...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on geoscience and remote sensing 2021-03, Vol.59 (3), p.2598-2610
Main Authors: Kang, Jian, Fernandez-Beltran, Ruben, Duan, Puhong, Liu, Sicong, Plaza, Antonio J.
Format: Article
Language:English
Subjects:
Archives & records
Artificial neural networks
Complexity
Complexity theory
Deep learning (DL)
Domains
Embedding
Feature extraction
Geography
Image contrast
Land cover
Learning
Measurement
metric learning
Momentum
Neural networks
Remote sensing
remote sensing (RS)
scene characterization
self-supervised learning
Semantics
Tiles
Transfer learning
unsupervised learning
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Convolutional neural networks (CNNs) have achieved great success when characterizing remote sensing (RS) images. However, the lack of sufficient annotated data (together with the high complexity of the RS image domain) often makes supervised and transfer learning schemes limited from an operational perspective. Despite the fact that unsupervised methods can potentially relieve these limitations, they are frequently unable to effectively exploit relevant prior knowledge about the RS domain, which may eventually constrain their final performance. In order to address these challenges, this article presents a new unsupervised deep metric learning model, called spatially augmented momentum contrast (SauMoCo), which has been specially designed to characterize unlabeled RS scenes. Based on the first law of geography, the proposed approach defines spatial augmentation criteria to uncover semantic relationships among land cover tiles. Then, a queue of deep embeddings is constructed to enhance the semantic variety of RS tiles within the considered contrastive learning process, where an auxiliary CNN model serves as an updating mechanism. Our experimental comparison, including different state-of-the-art techniques and benchmark RS image archives, reveals that the proposed approach obtains remarkable performance gains when characterizing unlabeled scenes since it is able to substantially enhance the discrimination ability among complex land cover categories. The source codes of this article will be made available to the RS community for reproducible research.
ISSN:0196-2892
1558-0644
DOI:10.1109/TGRS.2020.3007029