Beets or Cotton? Blind Extraction of Fine Agricultural Classes Using a Convolutional Autoencoder Applied to Temporal SAR Signatures

We present a fully unsupervised learning pipeline, which involves both a projection method and a clustering algorithm dedicated to the pixel-wise classification of multitemporal SAR images. We design a Convolutional Autoencoder as the method to project our time series onto a lower dimensional latent...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on geoscience and remote sensing 2022-01, Vol.60, p.1-18
Main Authors: Di Martino, Thomas, Guinvarc'h, Regis, Thirion-Lefevre, Laetitia, Koeniguer, Elise Colin
Format: Article
Language:English
Subjects:
Agriculture
Algorithms
autoencoder
Clustering
Computer Science
Convolution
Engineering Sciences
Feature extraction
Image classification
Machine learning
Mathematics
Physics
SAR
Semantics
Submarine pipelines
Synthetic aperture radar
Task analysis
Time series
Time series analysis
unsupervised classification
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:We present a fully unsupervised learning pipeline, which involves both a projection method and a clustering algorithm dedicated to the pixel-wise classification of multitemporal SAR images. We design a Convolutional Autoencoder as the method to project our time series onto a lower dimensional latent space, where semantically similar temporal signals are placed close together. The additional use of convolutional layers as feature extraction steps allows us to exploit the sequential nature of time series, exhibiting higher representation performance than fully connected layers. The extracted clusters can encapture different semantic levels to either separate classes or extract outlying temporal signals. The application of this method to crop-types mapping enables the extraction of major crop-types within a scene, without supervision. In a labeled context, this method also allows for the extraction of outlying profiles which can lead to the discovery of mislabeled time series.
ISSN:0196-2892
1558-0644
DOI:10.1109/TGRS.2021.3100637