Home > Research > Publications & Outputs > Land cover classification from remote sensing i...

Electronic data

  • 10095020.2021

    Rights statement: This is an Accepted Manuscript of an article published by Taylor & Francis in Geo-spatial Information Science on 07/01/2022, available online: http://www.tandfonline.com/10.1080/10095020.2021.2017237

    Accepted author manuscript, 16.9 MB, PDF document

    Available under license: CC BY: Creative Commons Attribution 4.0 International License

Links

Text available via DOI:

View graph of relations

Land cover classification from remote sensing images based on multi-scale fully convolutional network

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Published

Standard

Land cover classification from remote sensing images based on multi-scale fully convolutional network. / Li, Rui; Zheng, Shunyi; Duan, Chenxi et al.

In: Geo-spatial Information Science, Vol. 25, No. 2, 01.07.2022, p. 278-294.

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Harvard

Li, R, Zheng, S, Duan, C, Wang, L & Zhang, C 2022, 'Land cover classification from remote sensing images based on multi-scale fully convolutional network', Geo-spatial Information Science, vol. 25, no. 2, pp. 278-294. https://doi.org/10.1080/10095020.2021.2017237

APA

Vancouver

Li R, Zheng S, Duan C, Wang L, Zhang C. Land cover classification from remote sensing images based on multi-scale fully convolutional network. Geo-spatial Information Science. 2022 Jul 1;25(2):278-294. Epub 2022 Jan 7. doi: 10.1080/10095020.2021.2017237

Author

Li, Rui ; Zheng, Shunyi ; Duan, Chenxi et al. / Land cover classification from remote sensing images based on multi-scale fully convolutional network. In: Geo-spatial Information Science. 2022 ; Vol. 25, No. 2. pp. 278-294.

Bibtex

@article{fe4d233b84c44501a9607439f4eb5615,
title = "Land cover classification from remote sensing images based on multi-scale fully convolutional network",
abstract = "Although the Convolutional Neural Network (CNN) has shown great potential for land cover classification, the frequently used single-scale convolution kernel limits the scope of information extraction. Therefore, we propose a Multi-Scale Fully Convolutional Network (MSFCN) with a multi-scale convolutional kernel as well as a Channel Attention Block (CAB) and a Global Pooling Module (GPM) in this paper to exploit discriminative representations from two-dimensional (2D) satellite images. Meanwhile, to explore the ability of the proposed MSFCN for spatio-temporal images, we expand our MSFCN to three-dimension using three-dimensional (3D) CNN, capable of harnessing each land cover category{\textquoteright}s time series interaction from the reshaped spatio-temporal remote sensing images. To verify the effectiveness of the proposed MSFCN, we conduct experiments on two spatial datasets and two spatio-temporal datasets. The proposed MSFCN achieves 60.366% on the WHDLD dataset and 75.127% on the GID dataset in terms of mIoU index while the figures for two spatio-temporal datasets are 87.753% and 77.156%. Extensive comparative experiments and ablation studies demonstrate the effectiveness of the proposed MSFCN. Code will be available at https://github.com/lironui/MSFCN.",
keywords = "Spatio-temporal remote sensing images, Multi-Scale Fully Convolutional Network, land cover classification",
author = "Rui Li and Shunyi Zheng and Chenxi Duan and Libo Wang and Ce Zhang",
note = "This is an Accepted Manuscript of an article published by Taylor & Francis in Geo-spatial Information Science on 07/01/2022, available online: http://www.tandfonline.com/10.1080/10095020.2021.2017237",
year = "2022",
month = jul,
day = "1",
doi = "10.1080/10095020.2021.2017237",
language = "English",
volume = "25",
pages = "278--294",
journal = "Geo-spatial Information Science",
issn = "1009-5020",
publisher = "Taylor and Francis Ltd.",
number = "2",

}

RIS

TY - JOUR

T1 - Land cover classification from remote sensing images based on multi-scale fully convolutional network

AU - Li, Rui

AU - Zheng, Shunyi

AU - Duan, Chenxi

AU - Wang, Libo

AU - Zhang, Ce

N1 - This is an Accepted Manuscript of an article published by Taylor & Francis in Geo-spatial Information Science on 07/01/2022, available online: http://www.tandfonline.com/10.1080/10095020.2021.2017237

PY - 2022/7/1

Y1 - 2022/7/1

N2 - Although the Convolutional Neural Network (CNN) has shown great potential for land cover classification, the frequently used single-scale convolution kernel limits the scope of information extraction. Therefore, we propose a Multi-Scale Fully Convolutional Network (MSFCN) with a multi-scale convolutional kernel as well as a Channel Attention Block (CAB) and a Global Pooling Module (GPM) in this paper to exploit discriminative representations from two-dimensional (2D) satellite images. Meanwhile, to explore the ability of the proposed MSFCN for spatio-temporal images, we expand our MSFCN to three-dimension using three-dimensional (3D) CNN, capable of harnessing each land cover category’s time series interaction from the reshaped spatio-temporal remote sensing images. To verify the effectiveness of the proposed MSFCN, we conduct experiments on two spatial datasets and two spatio-temporal datasets. The proposed MSFCN achieves 60.366% on the WHDLD dataset and 75.127% on the GID dataset in terms of mIoU index while the figures for two spatio-temporal datasets are 87.753% and 77.156%. Extensive comparative experiments and ablation studies demonstrate the effectiveness of the proposed MSFCN. Code will be available at https://github.com/lironui/MSFCN.

AB - Although the Convolutional Neural Network (CNN) has shown great potential for land cover classification, the frequently used single-scale convolution kernel limits the scope of information extraction. Therefore, we propose a Multi-Scale Fully Convolutional Network (MSFCN) with a multi-scale convolutional kernel as well as a Channel Attention Block (CAB) and a Global Pooling Module (GPM) in this paper to exploit discriminative representations from two-dimensional (2D) satellite images. Meanwhile, to explore the ability of the proposed MSFCN for spatio-temporal images, we expand our MSFCN to three-dimension using three-dimensional (3D) CNN, capable of harnessing each land cover category’s time series interaction from the reshaped spatio-temporal remote sensing images. To verify the effectiveness of the proposed MSFCN, we conduct experiments on two spatial datasets and two spatio-temporal datasets. The proposed MSFCN achieves 60.366% on the WHDLD dataset and 75.127% on the GID dataset in terms of mIoU index while the figures for two spatio-temporal datasets are 87.753% and 77.156%. Extensive comparative experiments and ablation studies demonstrate the effectiveness of the proposed MSFCN. Code will be available at https://github.com/lironui/MSFCN.

KW - Spatio-temporal remote sensing images

KW - Multi-Scale Fully Convolutional Network

KW - land cover classification

U2 - 10.1080/10095020.2021.2017237

DO - 10.1080/10095020.2021.2017237

M3 - Journal article

VL - 25

SP - 278

EP - 294

JO - Geo-spatial Information Science

JF - Geo-spatial Information Science

SN - 1009-5020

IS - 2

ER -