MSMatch: Semisupervised Multispectral Scene Classification With Few Labels

Supervised learning techniques are at the center of many tasks in remote sensing. Unfortunately, these methods, especially recent deep learning methods, often require large amounts of labeled data for training. Even though satellites acquire large amounts of data, labeling the data is often tedious,...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Pablo Gomez, Gabriele Meoni
Formato:	article
Lenguaje:	EN
Publicado:	IEEE 2021
Materias:	Deep learning multispectral image classification scene classification semisupervised learning Ocean engineering TC1501-1800 Geophysics. Cosmic physics QC801-809
Acceso en línea:	https://doaj.org/article/9e2ab79646164b8dbff0a9b6431248bb
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:9e2ab79646164b8dbff0a9b6431248bb
record_format	dspace
spelling	oai:doaj.org-article:9e2ab79646164b8dbff0a9b6431248bb2021-11-25T00:00:10ZMSMatch: Semisupervised Multispectral Scene Classification With Few Labels2151-153510.1109/JSTARS.2021.3126082https://doaj.org/article/9e2ab79646164b8dbff0a9b6431248bb2021-01-01T00:00:00Zhttps://ieeexplore.ieee.org/document/9609561/https://doaj.org/toc/2151-1535Supervised learning techniques are at the center of many tasks in remote sensing. Unfortunately, these methods, especially recent deep learning methods, often require large amounts of labeled data for training. Even though satellites acquire large amounts of data, labeling the data is often tedious, expensive, and requires expert knowledge. Hence, improved methods that require fewer labeled samples are needed. We present MSMatch, the first semisupervised learning approach competitive with supervised methods on scene classification on the EuroSAT and UC Merced Land Use benchmark datasets. We test both RGB and multispectral images of EuroSAT and perform various ablation studies to identify the critical parts of the model. The trained neural network outperforms previous methods by up to 19.76% and 5.59% on EuroSAT and the UC Merced Land Use datasets, respectively. With just five labeled examples per class, we attain 90.71% and 95.86% accuracy on the UC Merced Land Use dataset and EuroSAT, respectively. Our results show that MSMatch is capable of greatly reducing the requirements for labeled data. It translates well to multispectral data and should enable various applications that are currently infeasible due to a lack of labeled data. We provide the source code of MSMatch online to enable easy reproduction and quick adoption.Pablo GomezGabriele MeoniIEEEarticleDeep learningmultispectral image classificationscene classificationsemisupervised learningOcean engineeringTC1501-1800Geophysics. Cosmic physicsQC801-809ENIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, Vol 14, Pp 11643-11654 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Deep learning multispectral image classification scene classification semisupervised learning Ocean engineering TC1501-1800 Geophysics. Cosmic physics QC801-809
spellingShingle	Deep learning multispectral image classification scene classification semisupervised learning Ocean engineering TC1501-1800 Geophysics. Cosmic physics QC801-809 Pablo Gomez Gabriele Meoni MSMatch: Semisupervised Multispectral Scene Classification With Few Labels
description	Supervised learning techniques are at the center of many tasks in remote sensing. Unfortunately, these methods, especially recent deep learning methods, often require large amounts of labeled data for training. Even though satellites acquire large amounts of data, labeling the data is often tedious, expensive, and requires expert knowledge. Hence, improved methods that require fewer labeled samples are needed. We present MSMatch, the first semisupervised learning approach competitive with supervised methods on scene classification on the EuroSAT and UC Merced Land Use benchmark datasets. We test both RGB and multispectral images of EuroSAT and perform various ablation studies to identify the critical parts of the model. The trained neural network outperforms previous methods by up to 19.76% and 5.59% on EuroSAT and the UC Merced Land Use datasets, respectively. With just five labeled examples per class, we attain 90.71% and 95.86% accuracy on the UC Merced Land Use dataset and EuroSAT, respectively. Our results show that MSMatch is capable of greatly reducing the requirements for labeled data. It translates well to multispectral data and should enable various applications that are currently infeasible due to a lack of labeled data. We provide the source code of MSMatch online to enable easy reproduction and quick adoption.
format	article
author	Pablo Gomez Gabriele Meoni
author_facet	Pablo Gomez Gabriele Meoni
author_sort	Pablo Gomez
title	MSMatch: Semisupervised Multispectral Scene Classification With Few Labels
title_short	MSMatch: Semisupervised Multispectral Scene Classification With Few Labels
title_full	MSMatch: Semisupervised Multispectral Scene Classification With Few Labels
title_fullStr	MSMatch: Semisupervised Multispectral Scene Classification With Few Labels
title_full_unstemmed	MSMatch: Semisupervised Multispectral Scene Classification With Few Labels
title_sort	msmatch: semisupervised multispectral scene classification with few labels
publisher	IEEE
publishDate	2021
url	https://doaj.org/article/9e2ab79646164b8dbff0a9b6431248bb
work_keys_str_mv	AT pablogomez msmatchsemisupervisedmultispectralsceneclassificationwithfewlabels AT gabrielemeoni msmatchsemisupervisedmultispectralsceneclassificationwithfewlabels
_version_	1718414710205841408

MSMatch: Semisupervised Multispectral Scene Classification With Few Labels

Ejemplares similares