Global Context-Based Multilevel Feature Fusion Networks for Multilabel Remote Sensing Image Scene Classification

Different from the traditional remote sensing (RS) scene classification which uses a single scene label to holistically annotate an image, multilabel RS image classification uses a series of object labels to interpret a scene more deeply. For multilabel RS scene classification, there exist two vital...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Xin Wang, Lin Duan, Chen Ning
Formato:	article
Lenguaje:	EN
Publicado:	IEEE 2021
Materias:	Deep learning (DL) multilabel classification (MLC) multilevel fusion remote sensing (RS) Ocean engineering TC1501-1800 Geophysics. Cosmic physics QC801-809
Acceso en línea:	https://doaj.org/article/e491c9abdec14b8588818292887fcc29
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:e491c9abdec14b8588818292887fcc29
record_format	dspace
spelling	oai:doaj.org-article:e491c9abdec14b8588818292887fcc292021-11-18T00:00:33ZGlobal Context-Based Multilevel Feature Fusion Networks for Multilabel Remote Sensing Image Scene Classification2151-153510.1109/JSTARS.2021.3122464https://doaj.org/article/e491c9abdec14b8588818292887fcc292021-01-01T00:00:00Zhttps://ieeexplore.ieee.org/document/9585651/https://doaj.org/toc/2151-1535Different from the traditional remote sensing (RS) scene classification which uses a single scene label to holistically annotate an image, multilabel RS image classification uses a series of object labels to interpret a scene more deeply. For multilabel RS scene classification, there exist two vital problems. First, the objects with different semantic labels have smaller sizes and more scattered arrangements compared to backgrounds, making meaningful semantic feature extraction and representation severely hard. Second, an RS scene usually contains various kinds of objects, leading to exponential magnification of output label space size with the increase of the number of object categories. To simultaneously solve the challenges in features as well as label space and produce significant performance improvements, this article proposes a novel end-to-end deep learning architecture, which we term the global context-based multilevel feature fusion network. We verify the whole framework by conducting a great number of experiments on two publicly available multilabel datasets, and we also provide an ablation study exploring different modules inclusion in the framework. Experimental results demonstrate that the proposed method is superior to some popular networks for multilabel RS image scene classification.Xin WangLin DuanChen NingIEEEarticleDeep learning (DL)multilabel classification (MLC)multilevel fusionremote sensing (RS)Ocean engineeringTC1501-1800Geophysics. Cosmic physicsQC801-809ENIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, Vol 14, Pp 11179-11196 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Deep learning (DL) multilabel classification (MLC) multilevel fusion remote sensing (RS) Ocean engineering TC1501-1800 Geophysics. Cosmic physics QC801-809
spellingShingle	Deep learning (DL) multilabel classification (MLC) multilevel fusion remote sensing (RS) Ocean engineering TC1501-1800 Geophysics. Cosmic physics QC801-809 Xin Wang Lin Duan Chen Ning Global Context-Based Multilevel Feature Fusion Networks for Multilabel Remote Sensing Image Scene Classification
description	Different from the traditional remote sensing (RS) scene classification which uses a single scene label to holistically annotate an image, multilabel RS image classification uses a series of object labels to interpret a scene more deeply. For multilabel RS scene classification, there exist two vital problems. First, the objects with different semantic labels have smaller sizes and more scattered arrangements compared to backgrounds, making meaningful semantic feature extraction and representation severely hard. Second, an RS scene usually contains various kinds of objects, leading to exponential magnification of output label space size with the increase of the number of object categories. To simultaneously solve the challenges in features as well as label space and produce significant performance improvements, this article proposes a novel end-to-end deep learning architecture, which we term the global context-based multilevel feature fusion network. We verify the whole framework by conducting a great number of experiments on two publicly available multilabel datasets, and we also provide an ablation study exploring different modules inclusion in the framework. Experimental results demonstrate that the proposed method is superior to some popular networks for multilabel RS image scene classification.
format	article
author	Xin Wang Lin Duan Chen Ning
author_facet	Xin Wang Lin Duan Chen Ning
author_sort	Xin Wang
title	Global Context-Based Multilevel Feature Fusion Networks for Multilabel Remote Sensing Image Scene Classification
title_short	Global Context-Based Multilevel Feature Fusion Networks for Multilabel Remote Sensing Image Scene Classification
title_full	Global Context-Based Multilevel Feature Fusion Networks for Multilabel Remote Sensing Image Scene Classification
title_fullStr	Global Context-Based Multilevel Feature Fusion Networks for Multilabel Remote Sensing Image Scene Classification
title_full_unstemmed	Global Context-Based Multilevel Feature Fusion Networks for Multilabel Remote Sensing Image Scene Classification
title_sort	global context-based multilevel feature fusion networks for multilabel remote sensing image scene classification
publisher	IEEE
publishDate	2021
url	https://doaj.org/article/e491c9abdec14b8588818292887fcc29
work_keys_str_mv	AT xinwang globalcontextbasedmultilevelfeaturefusionnetworksformultilabelremotesensingimagesceneclassification AT linduan globalcontextbasedmultilevelfeaturefusionnetworksformultilabelremotesensingimagesceneclassification AT chenning globalcontextbasedmultilevelfeaturefusionnetworksformultilabelremotesensingimagesceneclassification
_version_	1718425257061122048

Global Context-Based Multilevel Feature Fusion Networks for Multilabel Remote Sensing Image Scene Classification

Ejemplares similares