Cargando…

NAS-HRIS: Automatic Design and Architecture Search of Neural Network for Semantic Segmentation in Remote Sensing Images

The segmentation of high-resolution (HR) remote sensing images is very important in modern society, especially in the fields of industry, agriculture and urban modelling. Through the neural network, the machine can effectively and accurately extract the surface feature information. However, using th...

Descripción completa

Detalles Bibliográficos
Autores principales: Zhang, Mingwei, Jing, Weipeng, Lin, Jingbo, Fang, Nengzhen, Wei, Wei, Woźniak, Marcin, Damaševičius, Robertas
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2020
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7570751/
https://www.ncbi.nlm.nih.gov/pubmed/32947860
http://dx.doi.org/10.3390/s20185292
_version_ 1783597019589246976
author Zhang, Mingwei
Jing, Weipeng
Lin, Jingbo
Fang, Nengzhen
Wei, Wei
Woźniak, Marcin
Damaševičius, Robertas
author_facet Zhang, Mingwei
Jing, Weipeng
Lin, Jingbo
Fang, Nengzhen
Wei, Wei
Woźniak, Marcin
Damaševičius, Robertas
author_sort Zhang, Mingwei
collection PubMed
description The segmentation of high-resolution (HR) remote sensing images is very important in modern society, especially in the fields of industry, agriculture and urban modelling. Through the neural network, the machine can effectively and accurately extract the surface feature information. However, using the traditional deep learning methods requires plentiful efforts in order to find a robust architecture. In this paper, we introduce a neural network architecture search (NAS) method, called NAS-HRIS, which can automatically search neural network architecture on the dataset. The proposed method embeds a directed acyclic graph (DAG) into the search space and designs the differentiable searching process, which enables it to learn an end-to-end searching rule by using gradient descent optimization. It uses the Gumbel-Max trick to provide an efficient way when drawing samples from a non-continuous probability distribution, and it improves the efficiency of searching and reduces the memory consumption. Compared with other NAS, NAS-HRIS consumes less GPU memory without reducing the accuracy, which corresponds to a large amount of HR remote sensing imagery data. We have carried out experiments on the WHUBuilding dataset and achieved 90.44% MIoU. In order to fully demonstrate the feasibility of the method, we made a new urban Beijing Building dataset, and conducted experiments on satellite images and non-single source images, achieving better results than SegNet, U-Net and Deeplab v3+ models, while the computational complexity of our network architecture is much smaller.
format Online
Article
Text
id pubmed-7570751
institution National Center for Biotechnology Information
language English
publishDate 2020
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-75707512020-10-28 NAS-HRIS: Automatic Design and Architecture Search of Neural Network for Semantic Segmentation in Remote Sensing Images Zhang, Mingwei Jing, Weipeng Lin, Jingbo Fang, Nengzhen Wei, Wei Woźniak, Marcin Damaševičius, Robertas Sensors (Basel) Article The segmentation of high-resolution (HR) remote sensing images is very important in modern society, especially in the fields of industry, agriculture and urban modelling. Through the neural network, the machine can effectively and accurately extract the surface feature information. However, using the traditional deep learning methods requires plentiful efforts in order to find a robust architecture. In this paper, we introduce a neural network architecture search (NAS) method, called NAS-HRIS, which can automatically search neural network architecture on the dataset. The proposed method embeds a directed acyclic graph (DAG) into the search space and designs the differentiable searching process, which enables it to learn an end-to-end searching rule by using gradient descent optimization. It uses the Gumbel-Max trick to provide an efficient way when drawing samples from a non-continuous probability distribution, and it improves the efficiency of searching and reduces the memory consumption. Compared with other NAS, NAS-HRIS consumes less GPU memory without reducing the accuracy, which corresponds to a large amount of HR remote sensing imagery data. We have carried out experiments on the WHUBuilding dataset and achieved 90.44% MIoU. In order to fully demonstrate the feasibility of the method, we made a new urban Beijing Building dataset, and conducted experiments on satellite images and non-single source images, achieving better results than SegNet, U-Net and Deeplab v3+ models, while the computational complexity of our network architecture is much smaller. MDPI 2020-09-16 /pmc/articles/PMC7570751/ /pubmed/32947860 http://dx.doi.org/10.3390/s20185292 Text en © 2020 by the authors. 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 (http://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Zhang, Mingwei
Jing, Weipeng
Lin, Jingbo
Fang, Nengzhen
Wei, Wei
Woźniak, Marcin
Damaševičius, Robertas
NAS-HRIS: Automatic Design and Architecture Search of Neural Network for Semantic Segmentation in Remote Sensing Images
title NAS-HRIS: Automatic Design and Architecture Search of Neural Network for Semantic Segmentation in Remote Sensing Images
title_full NAS-HRIS: Automatic Design and Architecture Search of Neural Network for Semantic Segmentation in Remote Sensing Images
title_fullStr NAS-HRIS: Automatic Design and Architecture Search of Neural Network for Semantic Segmentation in Remote Sensing Images
title_full_unstemmed NAS-HRIS: Automatic Design and Architecture Search of Neural Network for Semantic Segmentation in Remote Sensing Images
title_short NAS-HRIS: Automatic Design and Architecture Search of Neural Network for Semantic Segmentation in Remote Sensing Images
title_sort nas-hris: automatic design and architecture search of neural network for semantic segmentation in remote sensing images
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7570751/
https://www.ncbi.nlm.nih.gov/pubmed/32947860
http://dx.doi.org/10.3390/s20185292
work_keys_str_mv AT zhangmingwei nashrisautomaticdesignandarchitecturesearchofneuralnetworkforsemanticsegmentationinremotesensingimages
AT jingweipeng nashrisautomaticdesignandarchitecturesearchofneuralnetworkforsemanticsegmentationinremotesensingimages
AT linjingbo nashrisautomaticdesignandarchitecturesearchofneuralnetworkforsemanticsegmentationinremotesensingimages
AT fangnengzhen nashrisautomaticdesignandarchitecturesearchofneuralnetworkforsemanticsegmentationinremotesensingimages
AT weiwei nashrisautomaticdesignandarchitecturesearchofneuralnetworkforsemanticsegmentationinremotesensingimages
AT wozniakmarcin nashrisautomaticdesignandarchitecturesearchofneuralnetworkforsemanticsegmentationinremotesensingimages
AT damaseviciusrobertas nashrisautomaticdesignandarchitecturesearchofneuralnetworkforsemanticsegmentationinremotesensingimages