SNLRUX++ for Building Extraction From High-Resolution Remote Sensing Images

Building extraction plays an important role in high-resolution remote sensing image processing, which can be used as the basis for urban planning and demographic analysis. In recent years, many powerful general semantic segmentation models have emerged, but these models often perform poorly when tra...

Full description

Bibliographic Details
Main Authors: Yanjing Lei, Jiamin Yu, Sixian Chan, Wei Wu, Xiaoying Liu
Format: Article
Language:English
Published: IEEE 2022-01-01
Series:IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Subjects:
Online Access:https://ieeexplore.ieee.org/document/9652051/
_version_ 1828922884402184192
author Yanjing Lei
Jiamin Yu
Sixian Chan
Wei Wu
Xiaoying Liu
author_facet Yanjing Lei
Jiamin Yu
Sixian Chan
Wei Wu
Xiaoying Liu
author_sort Yanjing Lei
collection DOAJ
description Building extraction plays an important role in high-resolution remote sensing image processing, which can be used as the basis for urban planning and demographic analysis. In recent years, many powerful general semantic segmentation models have emerged, but these models often perform poorly when transferred to remote sensing images because of the characteristics of remote sensing images. To this end, we propose a new deep learning network called Selective Nonlocal ResUNeXt++ (SNLRUX++) for building extraction. First, the cascaded multiscale feature fusion is proposed to transform the high-performance image classification network ResNeXt into the segmentation network ResUNeXt++. Second, selective nonlocal operation is designed to establish long-range dependencies while avoiding introducing excessive noise and computational effort. Finally, multiscale prediction is applied as deep supervision to accelerate training and convergence, and improves prediction performance of objects at different scales. The experimental results on two different remote sensing image datasets show the effectiveness and generalization ability of the proposed method.
first_indexed 2024-12-13T22:29:25Z
format Article
id doaj.art-2cf172d2b0314701abd9c6be49f69e92
institution Directory Open Access Journal
issn 2151-1535
language English
last_indexed 2024-12-13T22:29:25Z
publishDate 2022-01-01
publisher IEEE
record_format Article
series IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
spelling doaj.art-2cf172d2b0314701abd9c6be49f69e922022-12-21T23:29:07ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing2151-15352022-01-011540942110.1109/JSTARS.2021.31357059652051SNLRUX++ for Building Extraction From High-Resolution Remote Sensing ImagesYanjing Lei0Jiamin Yu1https://orcid.org/0000-0001-8915-7204Sixian Chan2https://orcid.org/0000-0001-8916-1174Wei Wu3Xiaoying Liu4College of Computer Science and Technology, Zhejiang University of Technology, Hangzhou, ChinaCollege of Computer Science and Technology, Zhejiang University of Technology, Hangzhou, ChinaCollege of Computer Science and Technology, Zhejiang University of Technology, Hangzhou, ChinaCollege of Computer Science and Technology, Zhejiang University of Technology, Hangzhou, ChinaCollege of Computer Science and Technology, Zhejiang University of Technology, Hangzhou, ChinaBuilding extraction plays an important role in high-resolution remote sensing image processing, which can be used as the basis for urban planning and demographic analysis. In recent years, many powerful general semantic segmentation models have emerged, but these models often perform poorly when transferred to remote sensing images because of the characteristics of remote sensing images. To this end, we propose a new deep learning network called Selective Nonlocal ResUNeXt++ (SNLRUX++) for building extraction. First, the cascaded multiscale feature fusion is proposed to transform the high-performance image classification network ResNeXt into the segmentation network ResUNeXt++. Second, selective nonlocal operation is designed to establish long-range dependencies while avoiding introducing excessive noise and computational effort. Finally, multiscale prediction is applied as deep supervision to accelerate training and convergence, and improves prediction performance of objects at different scales. The experimental results on two different remote sensing image datasets show the effectiveness and generalization ability of the proposed method.https://ieeexplore.ieee.org/document/9652051/Building extractionconvolution neural networkdeep learninghigh-resolution imageremote sensing
spellingShingle Yanjing Lei
Jiamin Yu
Sixian Chan
Wei Wu
Xiaoying Liu
SNLRUX++ for Building Extraction From High-Resolution Remote Sensing Images
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Building extraction
convolution neural network
deep learning
high-resolution image
remote sensing
title SNLRUX++ for Building Extraction From High-Resolution Remote Sensing Images
title_full SNLRUX++ for Building Extraction From High-Resolution Remote Sensing Images
title_fullStr SNLRUX++ for Building Extraction From High-Resolution Remote Sensing Images
title_full_unstemmed SNLRUX++ for Building Extraction From High-Resolution Remote Sensing Images
title_short SNLRUX++ for Building Extraction From High-Resolution Remote Sensing Images
title_sort snlrux for building extraction from high resolution remote sensing images
topic Building extraction
convolution neural network
deep learning
high-resolution image
remote sensing
url https://ieeexplore.ieee.org/document/9652051/
work_keys_str_mv AT yanjinglei snlruxforbuildingextractionfromhighresolutionremotesensingimages
AT jiaminyu snlruxforbuildingextractionfromhighresolutionremotesensingimages
AT sixianchan snlruxforbuildingextractionfromhighresolutionremotesensingimages
AT weiwu snlruxforbuildingextractionfromhighresolutionremotesensingimages
AT xiaoyingliu snlruxforbuildingextractionfromhighresolutionremotesensingimages