Cargando…

Research on Ground Object Classification Method of High Resolution Remote-Sensing Images Based on Improved DeeplabV3+

Ground-object classification using remote-sensing images of high resolution is widely used in land planning, ecological monitoring, and resource protection. Traditional image segmentation technology has poor effect on complex scenes in high-resolution remote-sensing images. In the field of deep lear...

Descripción completa

Detalles Bibliográficos
Autores principales: Fu, Junjie, Yi, Xiaomei, Wang, Guoying, Mo, Lufeng, Wu, Peng, Kapula, Kasanda Ernest
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2022
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9571339/
https://www.ncbi.nlm.nih.gov/pubmed/36236574
http://dx.doi.org/10.3390/s22197477
_version_ 1784810339814080512
author Fu, Junjie
Yi, Xiaomei
Wang, Guoying
Mo, Lufeng
Wu, Peng
Kapula, Kasanda Ernest
author_facet Fu, Junjie
Yi, Xiaomei
Wang, Guoying
Mo, Lufeng
Wu, Peng
Kapula, Kasanda Ernest
author_sort Fu, Junjie
collection PubMed
description Ground-object classification using remote-sensing images of high resolution is widely used in land planning, ecological monitoring, and resource protection. Traditional image segmentation technology has poor effect on complex scenes in high-resolution remote-sensing images. In the field of deep learning, some deep neural networks are being applied to high-resolution remote-sensing image segmentation. The DeeplabV3+ network is a deep neural network based on encoder-decoder architecture, which is commonly used to segment images with high precision. However, the segmentation accuracy of high-resolution remote-sensing images is poor, the number of network parameters is large, and the cost of training network is high. Therefore, this paper improves the DeeplabV3+ network. Firstly, MobileNetV2 network was used as the backbone feature-extraction network, and an attention-mechanism module was added after the feature-extraction module and the ASPP module to introduce focal loss balance. Our design has the following advantages: it enhances the ability of network to extract image features; it reduces network training costs; and it achieves better semantic segmentation accuracy. Experiments on high-resolution remote-sensing image datasets show that the mIou of the proposed method on WHDLD datasets is 64.76%, 4.24% higher than traditional DeeplabV3+ network mIou, and the mIou on CCF BDCI datasets is 64.58%. This is 5.35% higher than traditional DeeplabV3+ network mIou and outperforms traditional DeeplabV3+, U-NET, PSP-NET and MACU-net networks.
format Online
Article
Text
id pubmed-9571339
institution National Center for Biotechnology Information
language English
publishDate 2022
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-95713392022-10-17 Research on Ground Object Classification Method of High Resolution Remote-Sensing Images Based on Improved DeeplabV3+ Fu, Junjie Yi, Xiaomei Wang, Guoying Mo, Lufeng Wu, Peng Kapula, Kasanda Ernest Sensors (Basel) Article Ground-object classification using remote-sensing images of high resolution is widely used in land planning, ecological monitoring, and resource protection. Traditional image segmentation technology has poor effect on complex scenes in high-resolution remote-sensing images. In the field of deep learning, some deep neural networks are being applied to high-resolution remote-sensing image segmentation. The DeeplabV3+ network is a deep neural network based on encoder-decoder architecture, which is commonly used to segment images with high precision. However, the segmentation accuracy of high-resolution remote-sensing images is poor, the number of network parameters is large, and the cost of training network is high. Therefore, this paper improves the DeeplabV3+ network. Firstly, MobileNetV2 network was used as the backbone feature-extraction network, and an attention-mechanism module was added after the feature-extraction module and the ASPP module to introduce focal loss balance. Our design has the following advantages: it enhances the ability of network to extract image features; it reduces network training costs; and it achieves better semantic segmentation accuracy. Experiments on high-resolution remote-sensing image datasets show that the mIou of the proposed method on WHDLD datasets is 64.76%, 4.24% higher than traditional DeeplabV3+ network mIou, and the mIou on CCF BDCI datasets is 64.58%. This is 5.35% higher than traditional DeeplabV3+ network mIou and outperforms traditional DeeplabV3+, U-NET, PSP-NET and MACU-net networks. MDPI 2022-10-02 /pmc/articles/PMC9571339/ /pubmed/36236574 http://dx.doi.org/10.3390/s22197477 Text en © 2022 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Fu, Junjie
Yi, Xiaomei
Wang, Guoying
Mo, Lufeng
Wu, Peng
Kapula, Kasanda Ernest
Research on Ground Object Classification Method of High Resolution Remote-Sensing Images Based on Improved DeeplabV3+
title Research on Ground Object Classification Method of High Resolution Remote-Sensing Images Based on Improved DeeplabV3+
title_full Research on Ground Object Classification Method of High Resolution Remote-Sensing Images Based on Improved DeeplabV3+
title_fullStr Research on Ground Object Classification Method of High Resolution Remote-Sensing Images Based on Improved DeeplabV3+
title_full_unstemmed Research on Ground Object Classification Method of High Resolution Remote-Sensing Images Based on Improved DeeplabV3+
title_short Research on Ground Object Classification Method of High Resolution Remote-Sensing Images Based on Improved DeeplabV3+
title_sort research on ground object classification method of high resolution remote-sensing images based on improved deeplabv3+
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9571339/
https://www.ncbi.nlm.nih.gov/pubmed/36236574
http://dx.doi.org/10.3390/s22197477
work_keys_str_mv AT fujunjie researchongroundobjectclassificationmethodofhighresolutionremotesensingimagesbasedonimproveddeeplabv3
AT yixiaomei researchongroundobjectclassificationmethodofhighresolutionremotesensingimagesbasedonimproveddeeplabv3
AT wangguoying researchongroundobjectclassificationmethodofhighresolutionremotesensingimagesbasedonimproveddeeplabv3
AT molufeng researchongroundobjectclassificationmethodofhighresolutionremotesensingimagesbasedonimproveddeeplabv3
AT wupeng researchongroundobjectclassificationmethodofhighresolutionremotesensingimagesbasedonimproveddeeplabv3
AT kapulakasandaernest researchongroundobjectclassificationmethodofhighresolutionremotesensingimagesbasedonimproveddeeplabv3