Loading…
Cross-Domain Segmentation of Fundus Vessels Based on Feature Space Alignment
The accurate segmentation of fundus vessels plays a very important role in the detection and treatment of fundus diseases. With the rapid development of Convolutional Neural Networks (CNN), some CNN-based methods have been proposed for the segmentation of fundus vessels which show a good segmentatio...
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 | 289 |
| container_issue | |
| container_start_page | 284 |
| container_title | |
| container_volume | |
| creator | Wang, Huaixin Zeng, Qingshan Yang, Lei Liu, Yanhong Bian, Guibin |
| description | The accurate segmentation of fundus vessels plays a very important role in the detection and treatment of fundus diseases. With the rapid development of Convolutional Neural Networks (CNN), some CNN-based methods have been proposed for the segmentation of fundus vessels which show a good segmentation performance, but they rely on much well-annotated data sets. Aimed at this issue, based on a small number of annotated images, a new segmentation network is proposed in this paper to realize the segmentation of fundus vessels in the cross-domain. Two different high-level feature space are aligned and the Wasserstein distance is used to train the antagonistic networks. Experiments show that the proposed method could acquire a good segmentation performance on the public data sets of the DRIVE and STARE data sets. |
| doi_str_mv | 10.1109/ICMA49215.2020.9233568 |
| format | conference_proceeding |
| fullrecord | <record><control><sourceid>ieee_CHZPO</sourceid><recordid>TN_cdi_ieee_primary_9233568</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>9233568</ieee_id><sourcerecordid>9233568</sourcerecordid><originalsourceid>FETCH-LOGICAL-i118t-8d1aa770a006b8f631d7221e6522b4e6517c11a6937c1570859d464e29b406f33</originalsourceid><addsrcrecordid>eNotj81Kw0AURkdBsNY-gSDzAon3zkzmZxmj1ULERVXclUlzUyL5KZlk4ds3YldncfgOfIzdI8SI4B422VuqnMAkFiAgdkLKRNsLdoNGWNQKtb1ki9mLyCj1fc1WIfwAgAQEdLhgeTb0IURPfevrjm_p0FI3-rHuO95XfD115RT4F4VATeCPPlDJZ7UmP04D8e3R74mnTX3o_na37KryTaDVmUv2uX7-yF6j_P1lk6V5VCPaMbIlem8MeABd2EpLLI0QSDoRolAz0OwRvXZyZmLAJq5UWpFwhQJdSblkd__dmoh2x6Fu_fC7O5-XJ376TQw</addsrcrecordid><sourcetype>Publisher</sourcetype><iscdi>true</iscdi><recordtype>conference_proceeding</recordtype></control><display><type>conference_proceeding</type><title>Cross-Domain Segmentation of Fundus Vessels Based on Feature Space Alignment</title><source>IEEE Xplore All Conference Series</source><creator>Wang, Huaixin ; Zeng, Qingshan ; Yang, Lei ; Liu, Yanhong ; Bian, Guibin</creator><creatorcontrib>Wang, Huaixin ; Zeng, Qingshan ; Yang, Lei ; Liu, Yanhong ; Bian, Guibin</creatorcontrib><description>The accurate segmentation of fundus vessels plays a very important role in the detection and treatment of fundus diseases. With the rapid development of Convolutional Neural Networks (CNN), some CNN-based methods have been proposed for the segmentation of fundus vessels which show a good segmentation performance, but they rely on much well-annotated data sets. Aimed at this issue, based on a small number of annotated images, a new segmentation network is proposed in this paper to realize the segmentation of fundus vessels in the cross-domain. Two different high-level feature space are aligned and the Wasserstein distance is used to train the antagonistic networks. Experiments show that the proposed method could acquire a good segmentation performance on the public data sets of the DRIVE and STARE data sets.</description><identifier>EISSN: 2152-744X</identifier><identifier>EISBN: 1728164168</identifier><identifier>EISBN: 1728164141</identifier><identifier>EISBN: 9781728164168</identifier><identifier>EISBN: 9781728164144</identifier><identifier>DOI: 10.1109/ICMA49215.2020.9233568</identifier><language>eng</language><publisher>IEEE</publisher><subject>Antagonistic training ; Automation ; Conferences ; Convolutional neural networks ; Fundus vessels ; Image segmentation ; Mechatronics ; Network architecture ; Training data ; U-net network ; Wasserstein distance</subject><ispartof>2020 IEEE International Conference on Mechatronics and Automation (ICMA), 2020, p.284-289</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/9233568$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>309,310,780,784,789,790,23930,23931,25140,27925,54555,54932</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/9233568$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Wang, Huaixin</creatorcontrib><creatorcontrib>Zeng, Qingshan</creatorcontrib><creatorcontrib>Yang, Lei</creatorcontrib><creatorcontrib>Liu, Yanhong</creatorcontrib><creatorcontrib>Bian, Guibin</creatorcontrib><title>Cross-Domain Segmentation of Fundus Vessels Based on Feature Space Alignment</title><title>2020 IEEE International Conference on Mechatronics and Automation (ICMA)</title><addtitle>ICMA</addtitle><description>The accurate segmentation of fundus vessels plays a very important role in the detection and treatment of fundus diseases. With the rapid development of Convolutional Neural Networks (CNN), some CNN-based methods have been proposed for the segmentation of fundus vessels which show a good segmentation performance, but they rely on much well-annotated data sets. Aimed at this issue, based on a small number of annotated images, a new segmentation network is proposed in this paper to realize the segmentation of fundus vessels in the cross-domain. Two different high-level feature space are aligned and the Wasserstein distance is used to train the antagonistic networks. Experiments show that the proposed method could acquire a good segmentation performance on the public data sets of the DRIVE and STARE data sets.</description><subject>Antagonistic training</subject><subject>Automation</subject><subject>Conferences</subject><subject>Convolutional neural networks</subject><subject>Fundus vessels</subject><subject>Image segmentation</subject><subject>Mechatronics</subject><subject>Network architecture</subject><subject>Training data</subject><subject>U-net network</subject><subject>Wasserstein distance</subject><issn>2152-744X</issn><isbn>1728164168</isbn><isbn>1728164141</isbn><isbn>9781728164168</isbn><isbn>9781728164144</isbn><fulltext>true</fulltext><rsrctype>conference_proceeding</rsrctype><creationdate>2020</creationdate><recordtype>conference_proceeding</recordtype><sourceid>6IE</sourceid><recordid>eNotj81Kw0AURkdBsNY-gSDzAon3zkzmZxmj1ULERVXclUlzUyL5KZlk4ds3YldncfgOfIzdI8SI4B422VuqnMAkFiAgdkLKRNsLdoNGWNQKtb1ki9mLyCj1fc1WIfwAgAQEdLhgeTb0IURPfevrjm_p0FI3-rHuO95XfD115RT4F4VATeCPPlDJZ7UmP04D8e3R74mnTX3o_na37KryTaDVmUv2uX7-yF6j_P1lk6V5VCPaMbIlem8MeABd2EpLLI0QSDoRolAz0OwRvXZyZmLAJq5UWpFwhQJdSblkd__dmoh2x6Fu_fC7O5-XJ376TQw</recordid><startdate>20201013</startdate><enddate>20201013</enddate><creator>Wang, Huaixin</creator><creator>Zeng, Qingshan</creator><creator>Yang, Lei</creator><creator>Liu, Yanhong</creator><creator>Bian, Guibin</creator><general>IEEE</general><scope>6IE</scope><scope>6IL</scope><scope>CBEJK</scope><scope>RIE</scope><scope>RIL</scope></search><sort><creationdate>20201013</creationdate><title>Cross-Domain Segmentation of Fundus Vessels Based on Feature Space Alignment</title><author>Wang, Huaixin ; Zeng, Qingshan ; Yang, Lei ; Liu, Yanhong ; Bian, Guibin</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-i118t-8d1aa770a006b8f631d7221e6522b4e6517c11a6937c1570859d464e29b406f33</frbrgroupid><rsrctype>conference_proceedings</rsrctype><prefilter>conference_proceedings</prefilter><language>eng</language><creationdate>2020</creationdate><topic>Antagonistic training</topic><topic>Automation</topic><topic>Conferences</topic><topic>Convolutional neural networks</topic><topic>Fundus vessels</topic><topic>Image segmentation</topic><topic>Mechatronics</topic><topic>Network architecture</topic><topic>Training data</topic><topic>U-net network</topic><topic>Wasserstein distance</topic><toplevel>online_resources</toplevel><creatorcontrib>Wang, Huaixin</creatorcontrib><creatorcontrib>Zeng, Qingshan</creatorcontrib><creatorcontrib>Yang, Lei</creatorcontrib><creatorcontrib>Liu, Yanhong</creatorcontrib><creatorcontrib>Bian, Guibin</creatorcontrib><collection>IEEE Electronic Library (IEL) Conference Proceedings</collection><collection>IEEE Proceedings Order Plan All Online (POP All Online) 1998-present by volume</collection><collection>IEEE Xplore All Conference Proceedings</collection><collection>IEEE Electronic Library (IEL)</collection><collection>IEEE Proceedings Order Plans (POP All) 1998-Present</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Wang, Huaixin</au><au>Zeng, Qingshan</au><au>Yang, Lei</au><au>Liu, Yanhong</au><au>Bian, Guibin</au><format>book</format><genre>proceeding</genre><ristype>CONF</ristype><atitle>Cross-Domain Segmentation of Fundus Vessels Based on Feature Space Alignment</atitle><btitle>2020 IEEE International Conference on Mechatronics and Automation (ICMA)</btitle><stitle>ICMA</stitle><date>2020-10-13</date><risdate>2020</risdate><spage>284</spage><epage>289</epage><pages>284-289</pages><eissn>2152-744X</eissn><eisbn>1728164168</eisbn><eisbn>1728164141</eisbn><eisbn>9781728164168</eisbn><eisbn>9781728164144</eisbn><abstract>The accurate segmentation of fundus vessels plays a very important role in the detection and treatment of fundus diseases. With the rapid development of Convolutional Neural Networks (CNN), some CNN-based methods have been proposed for the segmentation of fundus vessels which show a good segmentation performance, but they rely on much well-annotated data sets. Aimed at this issue, based on a small number of annotated images, a new segmentation network is proposed in this paper to realize the segmentation of fundus vessels in the cross-domain. Two different high-level feature space are aligned and the Wasserstein distance is used to train the antagonistic networks. Experiments show that the proposed method could acquire a good segmentation performance on the public data sets of the DRIVE and STARE data sets.</abstract><pub>IEEE</pub><doi>10.1109/ICMA49215.2020.9233568</doi><tpages>6</tpages></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | EISSN: 2152-744X |
| ispartof | 2020 IEEE International Conference on Mechatronics and Automation (ICMA), 2020, p.284-289 |
| issn | 2152-744X |
| language | eng |
| recordid | cdi_ieee_primary_9233568 |
| source | IEEE Xplore All Conference Series |
| subjects | Antagonistic training Automation Conferences Convolutional neural networks Fundus vessels Image segmentation Mechatronics Network architecture Training data U-net network Wasserstein distance |
| title | Cross-Domain Segmentation of Fundus Vessels Based on Feature Space Alignment |
| url | http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-04T00%3A56%3A52IST&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=Cross-Domain%20Segmentation%20of%20Fundus%20Vessels%20Based%20on%20Feature%20Space%20Alignment&rft.btitle=2020%20IEEE%20International%20Conference%20on%20Mechatronics%20and%20Automation%20(ICMA)&rft.au=Wang,%20Huaixin&rft.date=2020-10-13&rft.spage=284&rft.epage=289&rft.pages=284-289&rft.eissn=2152-744X&rft_id=info:doi/10.1109/ICMA49215.2020.9233568&rft.eisbn=1728164168&rft.eisbn_list=1728164141&rft.eisbn_list=9781728164168&rft.eisbn_list=9781728164144&rft_dat=%3Cieee_CHZPO%3E9233568%3C/ieee_CHZPO%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-i118t-8d1aa770a006b8f631d7221e6522b4e6517c11a6937c1570859d464e29b406f33%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=9233568&rfr_iscdi=true |