Loading…
Examining Model Generality of Instance Segmentation for Building Mapping in Satellite Images - Case Study for Tokyo and Bangkok
In this study, we created a building extraction model from satellite images of Tokyo, which is updated more frequently than developing countries and has abundant data for training, and examined the possibility of extrapolating it to Bangkok, one of the megacities in developing countries. A deep lear...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Conference Proceeding |
| Language: | English |
| Subjects: | |
| Online Access: | Request full text |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| cited_by | |
|---|---|
| cites | |
| container_end_page | 5727 |
| container_issue | |
| container_start_page | 5724 |
| container_title | |
| container_volume | |
| creator | Yamanotera, Ryota Akiyama, Yuki Miyazaki, Hiroyuki |
| description | In this study, we created a building extraction model from satellite images of Tokyo, which is updated more frequently than developing countries and has abundant data for training, and examined the possibility of extrapolating it to Bangkok, one of the megacities in developing countries. A deep learning model using Meta's detectron2 library was used as the model for building extraction. The results of extrapolating to Bangkok with the model built in Tokyo showed that both IoU and Building Extraction Rate were more than 60% accurate. It was confirmed that this model can be extrapolated with the same performance as the model constructed in Bangkok for the extraction of buildings in Bangkok. This study also conducted a field survey in Bangkok to identify the causes of the obstacles to improving the accuracy of extrapolating the Tokyo model to Bangkok. The results revealed that the roof color and vegetation around the building, which are unique to Bangkok, affected the performance of the model. |
| doi_str_mv | 10.1109/IGARSS52108.2023.10282156 |
| format | conference_proceeding |
| fullrecord | <record><control><sourceid>ieee_CHZPO</sourceid><recordid>TN_cdi_ieee_primary_10282156</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>10282156</ieee_id><sourcerecordid>10282156</sourcerecordid><originalsourceid>FETCH-LOGICAL-i204t-34f9690a0da16869095ae3a41d4beab0ee35fb7a5a1cbd51a97a83248ae56b8d3</originalsourceid><addsrcrecordid>eNo1kMFKw0AURUdBsNb-gYvxA1LfzGSSybIttQYUwdR1eem8hLHJpCQpmJW_bqy6uocL5y4uY_cC5kJA8pBuFm9ZpqUAM5cg1VyANFLo6ILNkjgxSoOSICC-ZJOxVkEMoK7ZTdd9jGAkwIR9rT-xdt75kr80liq-IU8tVq4feFPw1Hc9-j3xjMqafI-9azwvmpYvT66yZw2Px590nmfYUzWqxNMaS-p4wFfYjXJ_ssPZ2jaHoeHoLV-iLw_N4ZZdFVh1NPvLKXt_XG9XT8Hz6yZdLZ4DJyHsAxUWSZQAgkURmZESjaQwFDbMCXMgUrrIY9Qo9rnVApMYjZKhQdJRbqyasrvfXUdEu2PramyH3f9h6hsmNGIS</addsrcrecordid><sourcetype>Publisher</sourcetype><iscdi>true</iscdi><recordtype>conference_proceeding</recordtype></control><display><type>conference_proceeding</type><title>Examining Model Generality of Instance Segmentation for Building Mapping in Satellite Images - Case Study for Tokyo and Bangkok</title><source>IEEE Xplore All Conference Series</source><creator>Yamanotera, Ryota ; Akiyama, Yuki ; Miyazaki, Hiroyuki</creator><creatorcontrib>Yamanotera, Ryota ; Akiyama, Yuki ; Miyazaki, Hiroyuki</creatorcontrib><description>In this study, we created a building extraction model from satellite images of Tokyo, which is updated more frequently than developing countries and has abundant data for training, and examined the possibility of extrapolating it to Bangkok, one of the megacities in developing countries. A deep learning model using Meta's detectron2 library was used as the model for building extraction. The results of extrapolating to Bangkok with the model built in Tokyo showed that both IoU and Building Extraction Rate were more than 60% accurate. It was confirmed that this model can be extrapolated with the same performance as the model constructed in Bangkok for the extraction of buildings in Bangkok. This study also conducted a field survey in Bangkok to identify the causes of the obstacles to improving the accuracy of extrapolating the Tokyo model to Bangkok. The results revealed that the roof color and vegetation around the building, which are unique to Bangkok, affected the performance of the model.</description><identifier>EISSN: 2153-7003</identifier><identifier>EISBN: 9798350320107</identifier><identifier>DOI: 10.1109/IGARSS52108.2023.10282156</identifier><language>eng</language><publisher>IEEE</publisher><subject>Building extraction ; Buildings ; Data models ; Deep learning ; Developing countries ; Satellite images ; Surveys ; Training ; Training data ; Urban planning ; Vegetation mapping</subject><ispartof>IGARSS 2023 - 2023 IEEE International Geoscience and Remote Sensing Symposium, 2023, p.5724-5727</ispartof><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/10282156$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>309,310,776,780,785,786,27904,54533,54910</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/10282156$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Yamanotera, Ryota</creatorcontrib><creatorcontrib>Akiyama, Yuki</creatorcontrib><creatorcontrib>Miyazaki, Hiroyuki</creatorcontrib><title>Examining Model Generality of Instance Segmentation for Building Mapping in Satellite Images - Case Study for Tokyo and Bangkok</title><title>IGARSS 2023 - 2023 IEEE International Geoscience and Remote Sensing Symposium</title><addtitle>IGARSS</addtitle><description>In this study, we created a building extraction model from satellite images of Tokyo, which is updated more frequently than developing countries and has abundant data for training, and examined the possibility of extrapolating it to Bangkok, one of the megacities in developing countries. A deep learning model using Meta's detectron2 library was used as the model for building extraction. The results of extrapolating to Bangkok with the model built in Tokyo showed that both IoU and Building Extraction Rate were more than 60% accurate. It was confirmed that this model can be extrapolated with the same performance as the model constructed in Bangkok for the extraction of buildings in Bangkok. This study also conducted a field survey in Bangkok to identify the causes of the obstacles to improving the accuracy of extrapolating the Tokyo model to Bangkok. The results revealed that the roof color and vegetation around the building, which are unique to Bangkok, affected the performance of the model.</description><subject>Building extraction</subject><subject>Buildings</subject><subject>Data models</subject><subject>Deep learning</subject><subject>Developing countries</subject><subject>Satellite images</subject><subject>Surveys</subject><subject>Training</subject><subject>Training data</subject><subject>Urban planning</subject><subject>Vegetation mapping</subject><issn>2153-7003</issn><isbn>9798350320107</isbn><fulltext>true</fulltext><rsrctype>conference_proceeding</rsrctype><creationdate>2023</creationdate><recordtype>conference_proceeding</recordtype><sourceid>6IE</sourceid><recordid>eNo1kMFKw0AURUdBsNb-gYvxA1LfzGSSybIttQYUwdR1eem8hLHJpCQpmJW_bqy6uocL5y4uY_cC5kJA8pBuFm9ZpqUAM5cg1VyANFLo6ILNkjgxSoOSICC-ZJOxVkEMoK7ZTdd9jGAkwIR9rT-xdt75kr80liq-IU8tVq4feFPw1Hc9-j3xjMqafI-9azwvmpYvT66yZw2Px590nmfYUzWqxNMaS-p4wFfYjXJ_ssPZ2jaHoeHoLV-iLw_N4ZZdFVh1NPvLKXt_XG9XT8Hz6yZdLZ4DJyHsAxUWSZQAgkURmZESjaQwFDbMCXMgUrrIY9Qo9rnVApMYjZKhQdJRbqyasrvfXUdEu2PramyH3f9h6hsmNGIS</recordid><startdate>20230716</startdate><enddate>20230716</enddate><creator>Yamanotera, Ryota</creator><creator>Akiyama, Yuki</creator><creator>Miyazaki, Hiroyuki</creator><general>IEEE</general><scope>6IE</scope><scope>6IH</scope><scope>CBEJK</scope><scope>RIE</scope><scope>RIO</scope></search><sort><creationdate>20230716</creationdate><title>Examining Model Generality of Instance Segmentation for Building Mapping in Satellite Images - Case Study for Tokyo and Bangkok</title><author>Yamanotera, Ryota ; Akiyama, Yuki ; Miyazaki, Hiroyuki</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-i204t-34f9690a0da16869095ae3a41d4beab0ee35fb7a5a1cbd51a97a83248ae56b8d3</frbrgroupid><rsrctype>conference_proceedings</rsrctype><prefilter>conference_proceedings</prefilter><language>eng</language><creationdate>2023</creationdate><topic>Building extraction</topic><topic>Buildings</topic><topic>Data models</topic><topic>Deep learning</topic><topic>Developing countries</topic><topic>Satellite images</topic><topic>Surveys</topic><topic>Training</topic><topic>Training data</topic><topic>Urban planning</topic><topic>Vegetation mapping</topic><toplevel>online_resources</toplevel><creatorcontrib>Yamanotera, Ryota</creatorcontrib><creatorcontrib>Akiyama, Yuki</creatorcontrib><creatorcontrib>Miyazaki, Hiroyuki</creatorcontrib><collection>IEEE Electronic Library (IEL) Conference Proceedings</collection><collection>IEEE Proceedings Order Plan (POP) 1998-present by volume</collection><collection>IEEE Xplore All Conference Proceedings</collection><collection>IEEE Electronic Library (IEL)</collection><collection>IEEE Proceedings Order Plans (POP) 1998-present</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Yamanotera, Ryota</au><au>Akiyama, Yuki</au><au>Miyazaki, Hiroyuki</au><format>book</format><genre>proceeding</genre><ristype>CONF</ristype><atitle>Examining Model Generality of Instance Segmentation for Building Mapping in Satellite Images - Case Study for Tokyo and Bangkok</atitle><btitle>IGARSS 2023 - 2023 IEEE International Geoscience and Remote Sensing Symposium</btitle><stitle>IGARSS</stitle><date>2023-07-16</date><risdate>2023</risdate><spage>5724</spage><epage>5727</epage><pages>5724-5727</pages><eissn>2153-7003</eissn><eisbn>9798350320107</eisbn><abstract>In this study, we created a building extraction model from satellite images of Tokyo, which is updated more frequently than developing countries and has abundant data for training, and examined the possibility of extrapolating it to Bangkok, one of the megacities in developing countries. A deep learning model using Meta's detectron2 library was used as the model for building extraction. The results of extrapolating to Bangkok with the model built in Tokyo showed that both IoU and Building Extraction Rate were more than 60% accurate. It was confirmed that this model can be extrapolated with the same performance as the model constructed in Bangkok for the extraction of buildings in Bangkok. This study also conducted a field survey in Bangkok to identify the causes of the obstacles to improving the accuracy of extrapolating the Tokyo model to Bangkok. The results revealed that the roof color and vegetation around the building, which are unique to Bangkok, affected the performance of the model.</abstract><pub>IEEE</pub><doi>10.1109/IGARSS52108.2023.10282156</doi><tpages>4</tpages></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | EISSN: 2153-7003 |
| ispartof | IGARSS 2023 - 2023 IEEE International Geoscience and Remote Sensing Symposium, 2023, p.5724-5727 |
| issn | 2153-7003 |
| language | eng |
| recordid | cdi_ieee_primary_10282156 |
| source | IEEE Xplore All Conference Series |
| subjects | Building extraction Buildings Data models Deep learning Developing countries Satellite images Surveys Training Training data Urban planning Vegetation mapping |
| title | Examining Model Generality of Instance Segmentation for Building Mapping in Satellite Images - Case Study for Tokyo and Bangkok |
| url | http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-26T23%3A04%3A16IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-ieee_CHZPO&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=proceeding&rft.atitle=Examining%20Model%20Generality%20of%20Instance%20Segmentation%20for%20Building%20Mapping%20in%20Satellite%20Images%20-%20Case%20Study%20for%20Tokyo%20and%20Bangkok&rft.btitle=IGARSS%202023%20-%202023%20IEEE%20International%20Geoscience%20and%20Remote%20Sensing%20Symposium&rft.au=Yamanotera,%20Ryota&rft.date=2023-07-16&rft.spage=5724&rft.epage=5727&rft.pages=5724-5727&rft.eissn=2153-7003&rft_id=info:doi/10.1109/IGARSS52108.2023.10282156&rft.eisbn=9798350320107&rft_dat=%3Cieee_CHZPO%3E10282156%3C/ieee_CHZPO%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-i204t-34f9690a0da16869095ae3a41d4beab0ee35fb7a5a1cbd51a97a83248ae56b8d3%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_id=info:pmid/&rft_ieee_id=10282156&rfr_iscdi=true |