Concatenated Residual Attention UNet for Semantic Segmentation of Urban Green Space

Urban green space is generally considered a significant component of the urban ecological environment system, which serves to improve the quality of the urban environment and provides various guarantees for the sustainable development of the city. Remote sensing provides an effective method for real...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Guoqiang Men, Guojin He, Guizhou Wang
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	urban green space remote sensing deep learning convolutional neural network residual structure attention mechanism Plant ecology QK900-989
Acceso en línea:	https://doaj.org/article/606841abfedd4df5a4ce5b303ae2b768
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:606841abfedd4df5a4ce5b303ae2b768
record_format	dspace
spelling	oai:doaj.org-article:606841abfedd4df5a4ce5b303ae2b7682021-11-25T17:37:19ZConcatenated Residual Attention UNet for Semantic Segmentation of Urban Green Space10.3390/f121114411999-4907https://doaj.org/article/606841abfedd4df5a4ce5b303ae2b7682021-10-01T00:00:00Zhttps://www.mdpi.com/1999-4907/12/11/1441https://doaj.org/toc/1999-4907Urban green space is generally considered a significant component of the urban ecological environment system, which serves to improve the quality of the urban environment and provides various guarantees for the sustainable development of the city. Remote sensing provides an effective method for real-time mapping and monitoring of urban green space changes in a large area. However, with the continuous improvement of the spatial resolution of remote sensing images, traditional classification methods cannot accurately obtain the spectral and spatial information of urban green spaces. Due to complex urban background and numerous shadows, there are mixed classifications for the extraction of cultivated land, grassland and other ground features, implying that limitations exist in traditional methods. At present, deep learning methods have shown great potential to tackle this challenge. In this research, we proposed a novel model called Concatenated Residual Attention UNet (CRAUNet), which combines the residual structure and channel attention mechanism, and applied it to the data source composed of GaoFen-1 remote sensing images in the Shenzhen City. Firstly, the improved residual structure is used to make it retain more feature information of the original image during the feature extraction process, then the Convolutional Block Channel Attention (CBCA) module is applied to enhance the extraction of deep convolution features by strengthening the effective green space features and suppressing invalid features through the interdependence of modeling channels.-Finally, the high-resolution feature map is restored through upsampling operation by the decoder. The experimental results show that compared with other methods, CRAUNet achieves the best performance. Especially, our method is less susceptible to the noise and preserves more complete segmented edge details. The pixel accuracy (PA) and mean intersection over union (MIoU) of our approach have reached 97.34% and 94.77%, which shows great applicability in regional large-scale mapping.Guoqiang MenGuojin HeGuizhou WangMDPI AGarticleurban green spaceremote sensingdeep learningconvolutional neural networkresidual structureattention mechanismPlant ecologyQK900-989ENForests, Vol 12, Iss 1441, p 1441 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	urban green space remote sensing deep learning convolutional neural network residual structure attention mechanism Plant ecology QK900-989
spellingShingle	urban green space remote sensing deep learning convolutional neural network residual structure attention mechanism Plant ecology QK900-989 Guoqiang Men Guojin He Guizhou Wang Concatenated Residual Attention UNet for Semantic Segmentation of Urban Green Space
description	Urban green space is generally considered a significant component of the urban ecological environment system, which serves to improve the quality of the urban environment and provides various guarantees for the sustainable development of the city. Remote sensing provides an effective method for real-time mapping and monitoring of urban green space changes in a large area. However, with the continuous improvement of the spatial resolution of remote sensing images, traditional classification methods cannot accurately obtain the spectral and spatial information of urban green spaces. Due to complex urban background and numerous shadows, there are mixed classifications for the extraction of cultivated land, grassland and other ground features, implying that limitations exist in traditional methods. At present, deep learning methods have shown great potential to tackle this challenge. In this research, we proposed a novel model called Concatenated Residual Attention UNet (CRAUNet), which combines the residual structure and channel attention mechanism, and applied it to the data source composed of GaoFen-1 remote sensing images in the Shenzhen City. Firstly, the improved residual structure is used to make it retain more feature information of the original image during the feature extraction process, then the Convolutional Block Channel Attention (CBCA) module is applied to enhance the extraction of deep convolution features by strengthening the effective green space features and suppressing invalid features through the interdependence of modeling channels.-Finally, the high-resolution feature map is restored through upsampling operation by the decoder. The experimental results show that compared with other methods, CRAUNet achieves the best performance. Especially, our method is less susceptible to the noise and preserves more complete segmented edge details. The pixel accuracy (PA) and mean intersection over union (MIoU) of our approach have reached 97.34% and 94.77%, which shows great applicability in regional large-scale mapping.
format	article
author	Guoqiang Men Guojin He Guizhou Wang
author_facet	Guoqiang Men Guojin He Guizhou Wang
author_sort	Guoqiang Men
title	Concatenated Residual Attention UNet for Semantic Segmentation of Urban Green Space
title_short	Concatenated Residual Attention UNet for Semantic Segmentation of Urban Green Space
title_full	Concatenated Residual Attention UNet for Semantic Segmentation of Urban Green Space
title_fullStr	Concatenated Residual Attention UNet for Semantic Segmentation of Urban Green Space
title_full_unstemmed	Concatenated Residual Attention UNet for Semantic Segmentation of Urban Green Space
title_sort	concatenated residual attention unet for semantic segmentation of urban green space
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/606841abfedd4df5a4ce5b303ae2b768
work_keys_str_mv	AT guoqiangmen concatenatedresidualattentionunetforsemanticsegmentationofurbangreenspace AT guojinhe concatenatedresidualattentionunetforsemanticsegmentationofurbangreenspace AT guizhouwang concatenatedresidualattentionunetforsemanticsegmentationofurbangreenspace
_version_	1718412167129071616

Concatenated Residual Attention UNet for Semantic Segmentation of Urban Green Space

Ejemplares similares