File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: Impervious surface estimation from optical and polarimetric sar data using small-patched deep convolutional networks: A comparative study

TitleImpervious surface estimation from optical and polarimetric sar data using small-patched deep convolutional networks: A comparative study
Authors
Keywordsconvolutional neural network (CNN)
deep learning
impervious surface area (ISA)
synthetic-aperture radar (SAR)
urban
Issue Date2019
Citation
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2019, v. 12, n. 7, p. 2374-2387 How to Cite?
Abstract© 2008-2012 IEEE. Incorporating optical and polarimetric synthetic-aperture radar (SAR) data to estimate impervious surface is useful but challenging due to their different geometric imaging mechanism and the high diversity of urban land covers. The recent development of deep convolutional networks (DCN) opens a promising opportunity by automatically extracting the deep features from both data sets. In this study, a small-patched DCN (SDCN) was designed to estimate the impervious surface from optical and SAR data. Benchmark methods, e.g., GoogLeNet, VGG16, ResNet50, and the support vector machine were employed for comparison. Two study sites in the most complex metropolitan of China, the Guangdong-Hong Kong-Macau Greater Bay Area, were selected to assess the proposed method. Experimental results indicated the effectiveness of proposed SDCN with a better accuracy outperforming other benchmark methods. Furthermore, we found that 60%-80% of training samples performed comparably with the whole training set, indicating that a large number of training samples may not be necessary in all cases, depending on the settings of some factors (e.g., number of epochs). Generally, SDCN appears more suitable than other methods in terms of combining the optical and SAR data and improved the accuracy of estimating impervious surface.
Persistent Identifierhttp://hdl.handle.net/10722/277708
ISSN
2017 Impact Factor: 2.777
2015 SCImago Journal Rankings: 1.196
ISI Accession Number ID

 

DC FieldValueLanguage
dc.contributor.authorZhang, Hongsheng-
dc.contributor.authorWan, Luoma-
dc.contributor.authorWang, Ting-
dc.contributor.authorLin, Yinyi-
dc.contributor.authorLin, Hui-
dc.contributor.authorZheng, Zezhong-
dc.date.accessioned2019-09-27T08:29:46Z-
dc.date.available2019-09-27T08:29:46Z-
dc.date.issued2019-
dc.identifier.citationIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2019, v. 12, n. 7, p. 2374-2387-
dc.identifier.issn1939-1404-
dc.identifier.urihttp://hdl.handle.net/10722/277708-
dc.description.abstract© 2008-2012 IEEE. Incorporating optical and polarimetric synthetic-aperture radar (SAR) data to estimate impervious surface is useful but challenging due to their different geometric imaging mechanism and the high diversity of urban land covers. The recent development of deep convolutional networks (DCN) opens a promising opportunity by automatically extracting the deep features from both data sets. In this study, a small-patched DCN (SDCN) was designed to estimate the impervious surface from optical and SAR data. Benchmark methods, e.g., GoogLeNet, VGG16, ResNet50, and the support vector machine were employed for comparison. Two study sites in the most complex metropolitan of China, the Guangdong-Hong Kong-Macau Greater Bay Area, were selected to assess the proposed method. Experimental results indicated the effectiveness of proposed SDCN with a better accuracy outperforming other benchmark methods. Furthermore, we found that 60%-80% of training samples performed comparably with the whole training set, indicating that a large number of training samples may not be necessary in all cases, depending on the settings of some factors (e.g., number of epochs). Generally, SDCN appears more suitable than other methods in terms of combining the optical and SAR data and improved the accuracy of estimating impervious surface.-
dc.languageeng-
dc.relation.ispartofIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing-
dc.subjectconvolutional neural network (CNN)-
dc.subjectdeep learning-
dc.subjectimpervious surface area (ISA)-
dc.subjectsynthetic-aperture radar (SAR)-
dc.subjecturban-
dc.titleImpervious surface estimation from optical and polarimetric sar data using small-patched deep convolutional networks: A comparative study-
dc.typeArticle-
dc.description.natureLink_to_subscribed_fulltext-
dc.identifier.doi10.1109/JSTARS.2019.2915277-
dc.identifier.scopuseid_2-s2.0-85070454847-
dc.identifier.volume12-
dc.identifier.issue7-
dc.identifier.spage2374-
dc.identifier.epage2387-
dc.identifier.eissn2151-1535-
dc.identifier.isiWOS:000480354800036-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats