Deep Fusion of DOM and DSM Features for Benggang Discovery

Benggang is a typical erosional landform in southern and southeastern China. Since benggang poses significant risks to local ecological environments and economic infrastructure, it is vital to accurately detect benggang-eroded areas. Relying only on remote sensing imagery for benggang detection cann...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:ISPRS international journal of geo-information 2021-08, Vol.10 (8), p.556, Article 556
Hauptverfasser: Shen, Shengyu, Chen, Jiasheng, Zhang, Shaoyi, Cheng, Dongbing, Wang, Zhigang, Zhang, Tong
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue 8
container_start_page 556
container_title ISPRS international journal of geo-information
container_volume 10
creator Shen, Shengyu
Chen, Jiasheng
Zhang, Shaoyi
Cheng, Dongbing
Wang, Zhigang
Zhang, Tong
description Benggang is a typical erosional landform in southern and southeastern China. Since benggang poses significant risks to local ecological environments and economic infrastructure, it is vital to accurately detect benggang-eroded areas. Relying only on remote sensing imagery for benggang detection cannot produce satisfactory results. In this study, we propose integrating high-resolution Digital Orthophoto Map (DOM) and Digital Surface Model (DSM) data for efficient and automatic benggang discovery. The fusion of complementary rich information hidden in both DOM and DSM data is realized by a two-stream convolutional neural network (CNN), which integrates aggregated terrain and activation image features that are both extracted by supervised deep learning. We aggregate local low-level geomorphic features via a supervised diffusion-convolutional embedding branch for expressive representations of benggang terrain variations. Activation image features are obtained from an image-oriented convolutional neural network branch. The two sources of information (DOM and DSM) are fused via a gated neural network, which learns the most discriminative features for the detection of benggang. The evaluation of a challenging benggang dataset demonstrates that our method exceeds several baselines, even with limited training examples. The results show that the fusion of DOM and DSM data is beneficial for benggang detection via supervised convolutional and deep fusion networks.
doi_str_mv 10.3390/ijgi10080556
format Article
fullrecord <record><control><sourceid>proquest_webof</sourceid><recordid>TN_cdi_webofscience_primary_000689272900001</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><doaj_id>oai_doaj_org_article_f6cafb7cb7374a818f0afc026a7bdda9</doaj_id><sourcerecordid>2565256484</sourcerecordid><originalsourceid>FETCH-LOGICAL-c367t-12c5251491474bdea3d794bfaa4a44d1eab011b48fe5456d954d5e1cbcf6acf13</originalsourceid><addsrcrecordid>eNqNkcFOJCEQhjtGE4168wFIPO6OAg10422dcdRE40E9kwKKDhO3GaF7N769PY4xHuVSlcpXf1X9VNUJo2d1rel5XHWRUdpSKdVOdcA5pzOtldj9lu9Xx6Ws6PQ0q1tBD6qLBeKaLMcSU09SIIuHewK9J4vHe7JEGMaMhYSUySX2XQd9RxaxuPQP89tRtRfgpeDxZzysnpdXT_Ob2d3D9e38z93M1aoZZow7ySUTmolGWI9Q-0YLGwAECOEZgqWMWdEGlEIqr6XwEpmzLihwgdWH1e1W1ydYmXWOfyG_mQTRfBRS7gzkIboXNEE5CLZxtqkbAS1rA4XgKFfQWO9BT1qnW611Tq8jlsGs0pj7aX3DpZr2VKIVE_V7S7mcSskYvqYyajZmm-9mT_ivLf4fbQrFRewdfrVMZqtW84brje-bc9qf0_M4wDB9zTyN_VC_A0HSkSQ</addsrcrecordid><sourcetype>Open Website</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2565256484</pqid></control><display><type>article</type><title>Deep Fusion of DOM and DSM Features for Benggang Discovery</title><source>DOAJ Directory of Open Access Journals</source><source>MDPI - Multidisciplinary Digital Publishing Institute</source><source>Web of Science - Science Citation Index Expanded - 2021&lt;img src="https://exlibris-pub.s3.amazonaws.com/fromwos-v2.jpg" /&gt;</source><source>EZB-FREE-00999 freely available EZB journals</source><creator>Shen, Shengyu ; Chen, Jiasheng ; Zhang, Shaoyi ; Cheng, Dongbing ; Wang, Zhigang ; Zhang, Tong</creator><creatorcontrib>Shen, Shengyu ; Chen, Jiasheng ; Zhang, Shaoyi ; Cheng, Dongbing ; Wang, Zhigang ; Zhang, Tong</creatorcontrib><description>Benggang is a typical erosional landform in southern and southeastern China. Since benggang poses significant risks to local ecological environments and economic infrastructure, it is vital to accurately detect benggang-eroded areas. Relying only on remote sensing imagery for benggang detection cannot produce satisfactory results. In this study, we propose integrating high-resolution Digital Orthophoto Map (DOM) and Digital Surface Model (DSM) data for efficient and automatic benggang discovery. The fusion of complementary rich information hidden in both DOM and DSM data is realized by a two-stream convolutional neural network (CNN), which integrates aggregated terrain and activation image features that are both extracted by supervised deep learning. We aggregate local low-level geomorphic features via a supervised diffusion-convolutional embedding branch for expressive representations of benggang terrain variations. Activation image features are obtained from an image-oriented convolutional neural network branch. The two sources of information (DOM and DSM) are fused via a gated neural network, which learns the most discriminative features for the detection of benggang. The evaluation of a challenging benggang dataset demonstrates that our method exceeds several baselines, even with limited training examples. The results show that the fusion of DOM and DSM data is beneficial for benggang detection via supervised convolutional and deep fusion networks.</description><identifier>ISSN: 2220-9964</identifier><identifier>EISSN: 2220-9964</identifier><identifier>DOI: 10.3390/ijgi10080556</identifier><language>eng</language><publisher>BASEL: Mdpi</publisher><subject>Artificial neural networks ; benggang ; CNN ; Computer Science ; Computer Science, Information Systems ; Datasets ; Deep learning ; Deformation ; Detection ; DOM ; DSM ; Economics ; Embedding ; Feature extraction ; fusion ; Geography, Physical ; Geomorphology ; Imagery ; Land degradation ; Landforms ; Landslides &amp; mudslides ; Machine learning ; Neural networks ; Physical Geography ; Physical Sciences ; Remote Sensing ; Science &amp; Technology ; Sustainable development ; Technology ; Terrain ; Training ; Unmanned aerial vehicles ; Vegetation</subject><ispartof>ISPRS international journal of geo-information, 2021-08, Vol.10 (8), p.556, Article 556</ispartof><rights>2021 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 (https://creativecommons.org/licenses/by/4.0/). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>true</woscitedreferencessubscribed><woscitedreferencescount>3</woscitedreferencescount><woscitedreferencesoriginalsourcerecordid>wos000689272900001</woscitedreferencesoriginalsourcerecordid><citedby>FETCH-LOGICAL-c367t-12c5251491474bdea3d794bfaa4a44d1eab011b48fe5456d954d5e1cbcf6acf13</citedby><cites>FETCH-LOGICAL-c367t-12c5251491474bdea3d794bfaa4a44d1eab011b48fe5456d954d5e1cbcf6acf13</cites><orcidid>0000-0002-0683-4669 ; 0000-0003-1573-4536 ; 0000-0001-6964-0354</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>315,781,785,865,2103,2115,27928,27929,39262</link.rule.ids></links><search><creatorcontrib>Shen, Shengyu</creatorcontrib><creatorcontrib>Chen, Jiasheng</creatorcontrib><creatorcontrib>Zhang, Shaoyi</creatorcontrib><creatorcontrib>Cheng, Dongbing</creatorcontrib><creatorcontrib>Wang, Zhigang</creatorcontrib><creatorcontrib>Zhang, Tong</creatorcontrib><title>Deep Fusion of DOM and DSM Features for Benggang Discovery</title><title>ISPRS international journal of geo-information</title><addtitle>ISPRS INT J GEO-INF</addtitle><description>Benggang is a typical erosional landform in southern and southeastern China. Since benggang poses significant risks to local ecological environments and economic infrastructure, it is vital to accurately detect benggang-eroded areas. Relying only on remote sensing imagery for benggang detection cannot produce satisfactory results. In this study, we propose integrating high-resolution Digital Orthophoto Map (DOM) and Digital Surface Model (DSM) data for efficient and automatic benggang discovery. The fusion of complementary rich information hidden in both DOM and DSM data is realized by a two-stream convolutional neural network (CNN), which integrates aggregated terrain and activation image features that are both extracted by supervised deep learning. We aggregate local low-level geomorphic features via a supervised diffusion-convolutional embedding branch for expressive representations of benggang terrain variations. Activation image features are obtained from an image-oriented convolutional neural network branch. The two sources of information (DOM and DSM) are fused via a gated neural network, which learns the most discriminative features for the detection of benggang. The evaluation of a challenging benggang dataset demonstrates that our method exceeds several baselines, even with limited training examples. The results show that the fusion of DOM and DSM data is beneficial for benggang detection via supervised convolutional and deep fusion networks.</description><subject>Artificial neural networks</subject><subject>benggang</subject><subject>CNN</subject><subject>Computer Science</subject><subject>Computer Science, Information Systems</subject><subject>Datasets</subject><subject>Deep learning</subject><subject>Deformation</subject><subject>Detection</subject><subject>DOM</subject><subject>DSM</subject><subject>Economics</subject><subject>Embedding</subject><subject>Feature extraction</subject><subject>fusion</subject><subject>Geography, Physical</subject><subject>Geomorphology</subject><subject>Imagery</subject><subject>Land degradation</subject><subject>Landforms</subject><subject>Landslides &amp; mudslides</subject><subject>Machine learning</subject><subject>Neural networks</subject><subject>Physical Geography</subject><subject>Physical Sciences</subject><subject>Remote Sensing</subject><subject>Science &amp; Technology</subject><subject>Sustainable development</subject><subject>Technology</subject><subject>Terrain</subject><subject>Training</subject><subject>Unmanned aerial vehicles</subject><subject>Vegetation</subject><issn>2220-9964</issn><issn>2220-9964</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2021</creationdate><recordtype>article</recordtype><sourceid>HGBXW</sourceid><sourceid>ABUWG</sourceid><sourceid>AFKRA</sourceid><sourceid>AZQEC</sourceid><sourceid>BENPR</sourceid><sourceid>CCPQU</sourceid><sourceid>DWQXO</sourceid><sourceid>DOA</sourceid><recordid>eNqNkcFOJCEQhjtGE4168wFIPO6OAg10422dcdRE40E9kwKKDhO3GaF7N769PY4xHuVSlcpXf1X9VNUJo2d1rel5XHWRUdpSKdVOdcA5pzOtldj9lu9Xx6Ws6PQ0q1tBD6qLBeKaLMcSU09SIIuHewK9J4vHe7JEGMaMhYSUySX2XQd9RxaxuPQP89tRtRfgpeDxZzysnpdXT_Ob2d3D9e38z93M1aoZZow7ySUTmolGWI9Q-0YLGwAECOEZgqWMWdEGlEIqr6XwEpmzLihwgdWH1e1W1ydYmXWOfyG_mQTRfBRS7gzkIboXNEE5CLZxtqkbAS1rA4XgKFfQWO9BT1qnW611Tq8jlsGs0pj7aX3DpZr2VKIVE_V7S7mcSskYvqYyajZmm-9mT_ivLf4fbQrFRewdfrVMZqtW84brje-bc9qf0_M4wDB9zTyN_VC_A0HSkSQ</recordid><startdate>20210801</startdate><enddate>20210801</enddate><creator>Shen, Shengyu</creator><creator>Chen, Jiasheng</creator><creator>Zhang, Shaoyi</creator><creator>Cheng, Dongbing</creator><creator>Wang, Zhigang</creator><creator>Zhang, Tong</creator><general>Mdpi</general><general>MDPI AG</general><scope>BLEPL</scope><scope>DTL</scope><scope>HGBXW</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7SC</scope><scope>7UA</scope><scope>8FD</scope><scope>8FE</scope><scope>8FG</scope><scope>ABJCF</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>ARAPS</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>BHPHI</scope><scope>BKSAR</scope><scope>C1K</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>F1W</scope><scope>FR3</scope><scope>H96</scope><scope>HCIFZ</scope><scope>JQ2</scope><scope>KR7</scope><scope>L.G</scope><scope>L6V</scope><scope>L7M</scope><scope>L~C</scope><scope>L~D</scope><scope>M7S</scope><scope>P5Z</scope><scope>P62</scope><scope>PCBAR</scope><scope>PIMPY</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PTHSS</scope><scope>DOA</scope><orcidid>https://orcid.org/0000-0002-0683-4669</orcidid><orcidid>https://orcid.org/0000-0003-1573-4536</orcidid><orcidid>https://orcid.org/0000-0001-6964-0354</orcidid></search><sort><creationdate>20210801</creationdate><title>Deep Fusion of DOM and DSM Features for Benggang Discovery</title><author>Shen, Shengyu ; Chen, Jiasheng ; Zhang, Shaoyi ; Cheng, Dongbing ; Wang, Zhigang ; Zhang, Tong</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c367t-12c5251491474bdea3d794bfaa4a44d1eab011b48fe5456d954d5e1cbcf6acf13</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2021</creationdate><topic>Artificial neural networks</topic><topic>benggang</topic><topic>CNN</topic><topic>Computer Science</topic><topic>Computer Science, Information Systems</topic><topic>Datasets</topic><topic>Deep learning</topic><topic>Deformation</topic><topic>Detection</topic><topic>DOM</topic><topic>DSM</topic><topic>Economics</topic><topic>Embedding</topic><topic>Feature extraction</topic><topic>fusion</topic><topic>Geography, Physical</topic><topic>Geomorphology</topic><topic>Imagery</topic><topic>Land degradation</topic><topic>Landforms</topic><topic>Landslides &amp; mudslides</topic><topic>Machine learning</topic><topic>Neural networks</topic><topic>Physical Geography</topic><topic>Physical Sciences</topic><topic>Remote Sensing</topic><topic>Science &amp; Technology</topic><topic>Sustainable development</topic><topic>Technology</topic><topic>Terrain</topic><topic>Training</topic><topic>Unmanned aerial vehicles</topic><topic>Vegetation</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Shen, Shengyu</creatorcontrib><creatorcontrib>Chen, Jiasheng</creatorcontrib><creatorcontrib>Zhang, Shaoyi</creatorcontrib><creatorcontrib>Cheng, Dongbing</creatorcontrib><creatorcontrib>Wang, Zhigang</creatorcontrib><creatorcontrib>Zhang, Tong</creatorcontrib><collection>Web of Science Core Collection</collection><collection>Science Citation Index Expanded</collection><collection>Web of Science - Science Citation Index Expanded - 2021</collection><collection>CrossRef</collection><collection>Computer and Information Systems Abstracts</collection><collection>Water Resources Abstracts</collection><collection>Technology Research Database</collection><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science &amp; Engineering Collection</collection><collection>ProQuest Central (Alumni Edition)</collection><collection>ProQuest Central UK/Ireland</collection><collection>Advanced Technologies &amp; Aerospace Collection</collection><collection>ProQuest Central Essentials</collection><collection>ProQuest Central</collection><collection>Technology Collection</collection><collection>Natural Science Collection</collection><collection>Earth, Atmospheric &amp; Aquatic Science Collection</collection><collection>Environmental Sciences and Pollution Management</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central Korea</collection><collection>ASFA: Aquatic Sciences and Fisheries Abstracts</collection><collection>Engineering Research Database</collection><collection>Aquatic Science &amp; Fisheries Abstracts (ASFA) 2: Ocean Technology, Policy &amp; Non-Living Resources</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Computer Science Collection</collection><collection>Civil Engineering Abstracts</collection><collection>Aquatic Science &amp; Fisheries Abstracts (ASFA) Professional</collection><collection>ProQuest Engineering Collection</collection><collection>Advanced Technologies Database with Aerospace</collection><collection>Computer and Information Systems Abstracts – Academic</collection><collection>Computer and Information Systems Abstracts Professional</collection><collection>Engineering Database</collection><collection>Advanced Technologies &amp; Aerospace Database</collection><collection>ProQuest Advanced Technologies &amp; Aerospace Collection</collection><collection>Earth, Atmospheric &amp; Aquatic Science Database</collection><collection>Publicly Available Content Database</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>Engineering Collection</collection><collection>DOAJ Directory of Open Access Journals</collection><jtitle>ISPRS international journal of geo-information</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Shen, Shengyu</au><au>Chen, Jiasheng</au><au>Zhang, Shaoyi</au><au>Cheng, Dongbing</au><au>Wang, Zhigang</au><au>Zhang, Tong</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Deep Fusion of DOM and DSM Features for Benggang Discovery</atitle><jtitle>ISPRS international journal of geo-information</jtitle><stitle>ISPRS INT J GEO-INF</stitle><date>2021-08-01</date><risdate>2021</risdate><volume>10</volume><issue>8</issue><spage>556</spage><pages>556-</pages><artnum>556</artnum><issn>2220-9964</issn><eissn>2220-9964</eissn><abstract>Benggang is a typical erosional landform in southern and southeastern China. Since benggang poses significant risks to local ecological environments and economic infrastructure, it is vital to accurately detect benggang-eroded areas. Relying only on remote sensing imagery for benggang detection cannot produce satisfactory results. In this study, we propose integrating high-resolution Digital Orthophoto Map (DOM) and Digital Surface Model (DSM) data for efficient and automatic benggang discovery. The fusion of complementary rich information hidden in both DOM and DSM data is realized by a two-stream convolutional neural network (CNN), which integrates aggregated terrain and activation image features that are both extracted by supervised deep learning. We aggregate local low-level geomorphic features via a supervised diffusion-convolutional embedding branch for expressive representations of benggang terrain variations. Activation image features are obtained from an image-oriented convolutional neural network branch. The two sources of information (DOM and DSM) are fused via a gated neural network, which learns the most discriminative features for the detection of benggang. The evaluation of a challenging benggang dataset demonstrates that our method exceeds several baselines, even with limited training examples. The results show that the fusion of DOM and DSM data is beneficial for benggang detection via supervised convolutional and deep fusion networks.</abstract><cop>BASEL</cop><pub>Mdpi</pub><doi>10.3390/ijgi10080556</doi><tpages>16</tpages><orcidid>https://orcid.org/0000-0002-0683-4669</orcidid><orcidid>https://orcid.org/0000-0003-1573-4536</orcidid><orcidid>https://orcid.org/0000-0001-6964-0354</orcidid><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier ISSN: 2220-9964
ispartof ISPRS international journal of geo-information, 2021-08, Vol.10 (8), p.556, Article 556
issn 2220-9964
2220-9964
language eng
recordid cdi_webofscience_primary_000689272900001
source DOAJ Directory of Open Access Journals; MDPI - Multidisciplinary Digital Publishing Institute; Web of Science - Science Citation Index Expanded - 2021<img src="https://exlibris-pub.s3.amazonaws.com/fromwos-v2.jpg" />; EZB-FREE-00999 freely available EZB journals
subjects Artificial neural networks
benggang
CNN
Computer Science
Computer Science, Information Systems
Datasets
Deep learning
Deformation
Detection
DOM
DSM
Economics
Embedding
Feature extraction
fusion
Geography, Physical
Geomorphology
Imagery
Land degradation
Landforms
Landslides & mudslides
Machine learning
Neural networks
Physical Geography
Physical Sciences
Remote Sensing
Science & Technology
Sustainable development
Technology
Terrain
Training
Unmanned aerial vehicles
Vegetation
title Deep Fusion of DOM and DSM Features for Benggang Discovery
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-16T16%3A31%3A57IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_webof&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Deep%20Fusion%20of%20DOM%20and%20DSM%20Features%20for%20Benggang%20Discovery&rft.jtitle=ISPRS%20international%20journal%20of%20geo-information&rft.au=Shen,%20Shengyu&rft.date=2021-08-01&rft.volume=10&rft.issue=8&rft.spage=556&rft.pages=556-&rft.artnum=556&rft.issn=2220-9964&rft.eissn=2220-9964&rft_id=info:doi/10.3390/ijgi10080556&rft_dat=%3Cproquest_webof%3E2565256484%3C/proquest_webof%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2565256484&rft_id=info:pmid/&rft_doaj_id=oai_doaj_org_article_f6cafb7cb7374a818f0afc026a7bdda9&rfr_iscdi=true