Fusion of semantic and appearance features for loop‐closure detection in a dynamic environment

Abstract Loop‐closure detection is important in large‐scale localisation system. However, it is still difficult in a dynamic environment. An online fast loop‐closure detection algorithm based on Deeplabv3 with MobileNetV2 (DpMn2) as network backbone and local difference binary descriptor is proposed...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Yan Xu, Jiani Huang
Formato:	article
Lenguaje:	EN
Publicado:	Wiley 2021
Materias:	Electrical engineering. Electronics. Nuclear engineering TK1-9971
Acceso en línea:	https://doaj.org/article/91e8951d056d47b99ba61deb0c0e8171
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:91e8951d056d47b99ba61deb0c0e8171
record_format	dspace
spelling	oai:doaj.org-article:91e8951d056d47b99ba61deb0c0e81712021-11-16T10:15:44ZFusion of semantic and appearance features for loop‐closure detection in a dynamic environment1350-911X0013-519410.1049/ell2.12052https://doaj.org/article/91e8951d056d47b99ba61deb0c0e81712021-01-01T00:00:00Zhttps://doi.org/10.1049/ell2.12052https://doaj.org/toc/0013-5194https://doaj.org/toc/1350-911XAbstract Loop‐closure detection is important in large‐scale localisation system. However, it is still difficult in a dynamic environment. An online fast loop‐closure detection algorithm based on Deeplabv3 with MobileNetV2 (DpMn2) as network backbone and local difference binary descriptor is proposed, and the algorithm is named as DpMn2‐LDB. DpMn2 splits out common dynamic objects of images, and then uses visual geometry group network (VGG16) that is trained on place‐centric data to extract global features for nearest neighbour image retrieval. The loop‐closure matches are verified based on LDB descriptors and random sample consensus (RANSAC). Experimental results show that the proposed method can obtain a higher recall rate under 100% precision with less execution time per frame on several public datasets compared with other typical or state‐of‐the‐art algorithms.Yan XuJiani HuangWileyarticleElectrical engineering. Electronics. Nuclear engineeringTK1-9971ENElectronics Letters, Vol 57, Iss 2, Pp 56-58 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Electrical engineering. Electronics. Nuclear engineering TK1-9971
spellingShingle	Electrical engineering. Electronics. Nuclear engineering TK1-9971 Yan Xu Jiani Huang Fusion of semantic and appearance features for loop‐closure detection in a dynamic environment
description	Abstract Loop‐closure detection is important in large‐scale localisation system. However, it is still difficult in a dynamic environment. An online fast loop‐closure detection algorithm based on Deeplabv3 with MobileNetV2 (DpMn2) as network backbone and local difference binary descriptor is proposed, and the algorithm is named as DpMn2‐LDB. DpMn2 splits out common dynamic objects of images, and then uses visual geometry group network (VGG16) that is trained on place‐centric data to extract global features for nearest neighbour image retrieval. The loop‐closure matches are verified based on LDB descriptors and random sample consensus (RANSAC). Experimental results show that the proposed method can obtain a higher recall rate under 100% precision with less execution time per frame on several public datasets compared with other typical or state‐of‐the‐art algorithms.
format	article
author	Yan Xu Jiani Huang
author_facet	Yan Xu Jiani Huang
author_sort	Yan Xu
title	Fusion of semantic and appearance features for loop‐closure detection in a dynamic environment
title_short	Fusion of semantic and appearance features for loop‐closure detection in a dynamic environment
title_full	Fusion of semantic and appearance features for loop‐closure detection in a dynamic environment
title_fullStr	Fusion of semantic and appearance features for loop‐closure detection in a dynamic environment
title_full_unstemmed	Fusion of semantic and appearance features for loop‐closure detection in a dynamic environment
title_sort	fusion of semantic and appearance features for loop‐closure detection in a dynamic environment
publisher	Wiley
publishDate	2021
url	https://doaj.org/article/91e8951d056d47b99ba61deb0c0e8171
work_keys_str_mv	AT yanxu fusionofsemanticandappearancefeaturesforloopclosuredetectioninadynamicenvironment AT jianihuang fusionofsemanticandappearancefeaturesforloopclosuredetectioninadynamicenvironment
_version_	1718426562269806592

Fusion of semantic and appearance features for loop‐closure detection in a dynamic environment

Ejemplares similares