File Download
  Links for fulltext
     (May Require Subscription)
Supplementary

Article: A small-patched convolutional neural network for mangrove mapping at species level using high-resolution remote-sensing image

TitleA small-patched convolutional neural network for mangrove mapping at species level using high-resolution remote-sensing image
Authors
KeywordsCNNs
Small patch
mangrove species
Issue Date2019
Citation
Annals of GIS, 2019, v. 25, n. 1, p. 45-55 How to Cite?
Abstract© 2019, © 2019 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group. The understanding of mangrove forest structure and dynamics at species level is essential for mangrove conservation and management. To classify mangrove species, remote-sensing technologies provide a better way with high spatial resolution image. The spatial structure is usually viewed as effective complementary information for classification. However, it is still a challenge to design handcrafted features for mangrove species due to their non-structure texture. To leverage the advantage of convolutional neural networks (CNNs) in abstract feature exploration, a small patch-based CNN is proposed to overcome the requirement of fixed and large input which limits the applicability of CNNs to fringe mangrove forests. The function of down-sampling technology was substantially reduced to make deeper network for small input in our work. Meanwhile, the inception structure is used to exploit the multi-scale features of mangrove forests. Furthermore, the network is optimized with lesser convolution kernels and a single fully connected layer to reduce overfitting via reducing the training parameters. Compared to the features of grey level co-occurrence matrix with support vector machine, our proposed CNN shows better performance in classification accuracy. Moreover, the differences between mangrove species can be perceptive via CNN visualization, which offers better understanding of mangrove forests.
Persistent Identifierhttp://hdl.handle.net/10722/277700
ISSN
2023 Impact Factor: 2.7
2023 SCImago Journal Rankings: 0.923
ISI Accession Number ID

 

DC FieldValueLanguage
dc.contributor.authorWan, Luoma-
dc.contributor.authorZhang, Hongsheng-
dc.contributor.authorLin, Guanghui-
dc.contributor.authorLin, Hui-
dc.date.accessioned2019-09-27T08:29:44Z-
dc.date.available2019-09-27T08:29:44Z-
dc.date.issued2019-
dc.identifier.citationAnnals of GIS, 2019, v. 25, n. 1, p. 45-55-
dc.identifier.issn1947-5683-
dc.identifier.urihttp://hdl.handle.net/10722/277700-
dc.description.abstract© 2019, © 2019 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group. The understanding of mangrove forest structure and dynamics at species level is essential for mangrove conservation and management. To classify mangrove species, remote-sensing technologies provide a better way with high spatial resolution image. The spatial structure is usually viewed as effective complementary information for classification. However, it is still a challenge to design handcrafted features for mangrove species due to their non-structure texture. To leverage the advantage of convolutional neural networks (CNNs) in abstract feature exploration, a small patch-based CNN is proposed to overcome the requirement of fixed and large input which limits the applicability of CNNs to fringe mangrove forests. The function of down-sampling technology was substantially reduced to make deeper network for small input in our work. Meanwhile, the inception structure is used to exploit the multi-scale features of mangrove forests. Furthermore, the network is optimized with lesser convolution kernels and a single fully connected layer to reduce overfitting via reducing the training parameters. Compared to the features of grey level co-occurrence matrix with support vector machine, our proposed CNN shows better performance in classification accuracy. Moreover, the differences between mangrove species can be perceptive via CNN visualization, which offers better understanding of mangrove forests.-
dc.languageeng-
dc.relation.ispartofAnnals of GIS-
dc.rightsThis work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.-
dc.subjectCNNs-
dc.subjectSmall patch-
dc.subjectmangrove species-
dc.titleA small-patched convolutional neural network for mangrove mapping at species level using high-resolution remote-sensing image-
dc.typeArticle-
dc.description.naturepublished_or_final_version-
dc.identifier.doi10.1080/19475683.2018.1564791-
dc.identifier.scopuseid_2-s2.0-85059888710-
dc.identifier.volume25-
dc.identifier.issue1-
dc.identifier.spage45-
dc.identifier.epage55-
dc.identifier.eissn1947-5691-
dc.identifier.isiWOS:000495707200005-
dc.identifier.issnl1947-5691-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats