Efficient compression of encoder-decoder models for semantic segmentation using the separation index

We present a novel approach to compressing encoder–decoder architectures, particularly in semantic segmentation tasks, by leveraging the Separation Index (SI)—a metric that quantifies how distinctly a network’s feature maps separate different classes at the pixel level. By identifying and pruning re...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Scientific reports Ročník 15; číslo 1; s. 24639 - 19
Hlavní autori: Jamshidi, Movahed, Kalhor, Ahmad, Vahabie, Abdol-Hossein
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: London Nature Publishing Group UK 09.07.2025
Nature Publishing Group
Nature Portfolio
Predmet:
ISSN:2045-2322, 2045-2322
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Abstract We present a novel approach to compressing encoder–decoder architectures, particularly in semantic segmentation tasks, by leveraging the Separation Index (SI)—a metric that quantifies how distinctly a network’s feature maps separate different classes at the pixel level. By identifying and pruning redundant layers and filters, our method preserves the fine-grained spatial details crucial for segmentation while significantly reducing model complexity. We evaluated our approach on five diverse datasets—CamVid (road scenes), KiTS19 (kidney tumor CT scans), the 2018 Data Science Bowl (nuclei segmentation), Aerial Imagery for remote sensing, and MVTec AD (industrial anomaly detection)—across architectures such as U-Net, LinkNet, MobileNet, DeepLabV3, and SegNet. Experimental results show that SI-driven compression reduces parameters and floating-point operations by up to 70% while maintaining or even improving segmentation accuracy, as measured by mean Intersection over Union (IoU). For example, a compressed DeepLabV3 raises the mean IoU from 0.624 to 0.638 on an aerial imagery dataset with a 2.6× reduction in parameters and faster inference. These findings highlight how SI-based pruning balances efficiency and performance, offering a practical solution for resource-constrained semantic segmentation applications.
AbstractList We present a novel approach to compressing encoder–decoder architectures, particularly in semantic segmentation tasks, by leveraging the Separation Index (SI)—a metric that quantifies how distinctly a network’s feature maps separate different classes at the pixel level. By identifying and pruning redundant layers and filters, our method preserves the fine-grained spatial details crucial for segmentation while significantly reducing model complexity. We evaluated our approach on five diverse datasets—CamVid (road scenes), KiTS19 (kidney tumor CT scans), the 2018 Data Science Bowl (nuclei segmentation), Aerial Imagery for remote sensing, and MVTec AD (industrial anomaly detection)—across architectures such as U-Net, LinkNet, MobileNet, DeepLabV3, and SegNet. Experimental results show that SI-driven compression reduces parameters and floating-point operations by up to 70% while maintaining or even improving segmentation accuracy, as measured by mean Intersection over Union (IoU). For example, a compressed DeepLabV3 raises the mean IoU from 0.624 to 0.638 on an aerial imagery dataset with a 2.6× reduction in parameters and faster inference. These findings highlight how SI-based pruning balances efficiency and performance, offering a practical solution for resource-constrained semantic segmentation applications.
Abstract We present a novel approach to compressing encoder–decoder architectures, particularly in semantic segmentation tasks, by leveraging the Separation Index (SI)—a metric that quantifies how distinctly a network’s feature maps separate different classes at the pixel level. By identifying and pruning redundant layers and filters, our method preserves the fine-grained spatial details crucial for segmentation while significantly reducing model complexity. We evaluated our approach on five diverse datasets—CamVid (road scenes), KiTS19 (kidney tumor CT scans), the 2018 Data Science Bowl (nuclei segmentation), Aerial Imagery for remote sensing, and MVTec AD (industrial anomaly detection)—across architectures such as U-Net, LinkNet, MobileNet, DeepLabV3, and SegNet. Experimental results show that SI-driven compression reduces parameters and floating-point operations by up to 70% while maintaining or even improving segmentation accuracy, as measured by mean Intersection over Union (IoU). For example, a compressed DeepLabV3 raises the mean IoU from 0.624 to 0.638 on an aerial imagery dataset with a 2.6× reduction in parameters and faster inference. These findings highlight how SI-based pruning balances efficiency and performance, offering a practical solution for resource-constrained semantic segmentation applications.
We present a novel approach to compressing encoder-decoder architectures, particularly in semantic segmentation tasks, by leveraging the Separation Index (SI)-a metric that quantifies how distinctly a network's feature maps separate different classes at the pixel level. By identifying and pruning redundant layers and filters, our method preserves the fine-grained spatial details crucial for segmentation while significantly reducing model complexity. We evaluated our approach on five diverse datasets-CamVid (road scenes), KiTS19 (kidney tumor CT scans), the 2018 Data Science Bowl (nuclei segmentation), Aerial Imagery for remote sensing, and MVTec AD (industrial anomaly detection)-across architectures such as U-Net, LinkNet, MobileNet, DeepLabV3, and SegNet. Experimental results show that SI-driven compression reduces parameters and floating-point operations by up to 70% while maintaining or even improving segmentation accuracy, as measured by mean Intersection over Union (IoU). For example, a compressed DeepLabV3 raises the mean IoU from 0.624 to 0.638 on an aerial imagery dataset with a 2.6× reduction in parameters and faster inference. These findings highlight how SI-based pruning balances efficiency and performance, offering a practical solution for resource-constrained semantic segmentation applications.We present a novel approach to compressing encoder-decoder architectures, particularly in semantic segmentation tasks, by leveraging the Separation Index (SI)-a metric that quantifies how distinctly a network's feature maps separate different classes at the pixel level. By identifying and pruning redundant layers and filters, our method preserves the fine-grained spatial details crucial for segmentation while significantly reducing model complexity. We evaluated our approach on five diverse datasets-CamVid (road scenes), KiTS19 (kidney tumor CT scans), the 2018 Data Science Bowl (nuclei segmentation), Aerial Imagery for remote sensing, and MVTec AD (industrial anomaly detection)-across architectures such as U-Net, LinkNet, MobileNet, DeepLabV3, and SegNet. Experimental results show that SI-driven compression reduces parameters and floating-point operations by up to 70% while maintaining or even improving segmentation accuracy, as measured by mean Intersection over Union (IoU). For example, a compressed DeepLabV3 raises the mean IoU from 0.624 to 0.638 on an aerial imagery dataset with a 2.6× reduction in parameters and faster inference. These findings highlight how SI-based pruning balances efficiency and performance, offering a practical solution for resource-constrained semantic segmentation applications.
ArticleNumber 24639
Author Vahabie, Abdol-Hossein
Kalhor, Ahmad
Jamshidi, Movahed
Author_xml – sequence: 1
  givenname: Movahed
  surname: Jamshidi
  fullname: Jamshidi, Movahed
  email: mo.jamshidi@ut.ac.ir
  organization: School of Electrical and Computer Engineering, University of Tehran
– sequence: 2
  givenname: Ahmad
  surname: Kalhor
  fullname: Kalhor, Ahmad
  organization: School of Electrical and Computer Engineering, University of Tehran
– sequence: 3
  givenname: Abdol-Hossein
  surname: Vahabie
  fullname: Vahabie, Abdol-Hossein
  organization: School of Electrical and Computer Engineering, University of Tehran
BackLink https://www.ncbi.nlm.nih.gov/pubmed/40634505$$D View this record in MEDLINE/PubMed
BookMark eNp9Uk1v1DAQtVARLaV_gAOKxIVLwB_xxj4hVBWoVIkLnC3HHm-9SuzFThD8e6abFtoe6oPHHr_39DwzL8lRygkIec3oe0aF-lA7JrVqKZct3jvV6mfkhNNOtlxwfnTvfEzOat1RXJLrjukX5LijG9FJKk-Ivwghughpblye9gVqjTk1OTSQXPZQWg-H2Ey4j7UJuTQVJpvm6PCwnZBq5xvOUmPaNvM1YHpvy5qMycPvV-R5sGOFs9t4Sn58vvh-_rW9-vbl8vzTVetkJ-a21z311DKvOyEp82wQQUlPFQAM1ts-BDsMQDVw50UfehVgwB_2QjK_2XhxSi5XXZ_tzuxLnGz5Y7KN5pDIZWtsQd8jGGDSKQ9Yo8A7poQCq4WSNrCeckcpan1ctfbLMIF3-M1ixweiD19SvDbb_MswjoKd5Kjw7lah5J8L1NlMsToYR5sgL9Vga5TiQtMeoW8fQXd5KQlrdUBtGFVaIurNfUv_vNx1EwFqBbiSay0QjItrc9BhHA2j5mZ2zDo7BmfHHGbHaKTyR9Q79SdJYiVVBKctlP-2n2D9BbMQ16g
CitedBy_id crossref_primary_10_1038_s41598_025_16604_2
Cites_doi 10.2139/ssrn.4796336
10.1109/ICDCECE60827.2024.10548323
10.1007/978-0-387-84858-7
10.1016/j.asoc.2024.112300
10.1109/CVPR.2019.00982
10.1109/ICSPIS56952.2022.10043890
10.1038/s41592-019-0612-7
10.1109/CVPRW50498.2020.00190
10.1109/ICCV.2019.00038
10.1109/BDCloud.2018.00110
10.1109/TPAMI.2017.2699184
10.1016/j.engappai.2024.108353
10.4103/jmss.jmss_108_21
10.1016/j.eswa.2023.119944
10.1109/TPAMI.2016.2644615
10.1109/TCYB.2019.2928174
10.3390/informatics8040077
10.1016/j.patrec.2008.04.005
10.1016/j.neucom.2024.127817
10.1007/978-3-319-24574-4_28
10.1109/MVIP62238.2024.10491151
10.1109/CVPR42600.2020.01318
10.1145/3140659.3080254
10.1016/j.neucom.2022.11.072
10.1109/ISCAS48785.2022.9937283
10.1109/72.248452
10.1109/IJCNN.1991.170743
10.1109/ACCESS.2020.3048375
10.3390/computers12030060
10.1109/CVPR.2019.00293
10.1109/TAC.1974.1100705
10.1109/ICCV.2017.155
10.1109/TIP.2003.819861
10.1364/BOE.395279
10.1007/978-3-030-76423-4_12
10.1007/s10489-022-03508-1
10.1109/CVPRW50498.2020.00088
10.1109/72.788640
10.1007/978-3-030-33642-4_13
10.1109/CVPR.2018.00810
10.1109/ICCV.2017.541
10.53941/ijndi.2024.100009
10.1145/3234804.3234824
10.1038/s41598-024-68172-6
10.1007/s10489-022-03229-5
10.1038/s41598-024-59176-3
10.1016/j.jksuci.2023.101560
ContentType Journal Article
Copyright The Author(s) 2025
2025. The Author(s).
The Author(s) 2025. This work is published under http://creativecommons.org/licenses/by-nc-nd/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
The Author(s) 2025 2025
Copyright_xml – notice: The Author(s) 2025
– notice: 2025. The Author(s).
– notice: The Author(s) 2025. This work is published under http://creativecommons.org/licenses/by-nc-nd/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
– notice: The Author(s) 2025 2025
DBID C6C
AAYXX
CITATION
NPM
3V.
7X7
7XB
88A
88E
88I
8FE
8FH
8FI
8FJ
8FK
ABUWG
AEUYN
AFKRA
AZQEC
BBNVY
BENPR
BHPHI
CCPQU
DWQXO
FYUFA
GHDGH
GNUQQ
HCIFZ
K9.
LK8
M0S
M1P
M2P
M7P
PHGZM
PHGZT
PIMPY
PJZUB
PKEHL
PPXIY
PQEST
PQGLB
PQQKQ
PQUKI
PRINS
Q9U
7X8
5PM
DOA
DOI 10.1038/s41598-025-10348-9
DatabaseName Springer Nature OA Free Journals
CrossRef
PubMed
ProQuest Central (Corporate)
Health & Medical Collection
ProQuest Central (purchase pre-March 2016)
Biology Database (Alumni Edition)
Medical Database (Alumni Edition)
Science Database (Alumni Edition)
ProQuest SciTech Collection
ProQuest Natural Science Collection
Hospital Premium Collection
Hospital Premium Collection (Alumni Edition)
ProQuest Central (Alumni) (purchase pre-March 2016)
ProQuest Central (Alumni)
ProQuest One Sustainability
ProQuest Central UK/Ireland
ProQuest Central Essentials
Biological Science Collection
ProQuest Central
Natural Science Collection
ProQuest One Community College
ProQuest Central
Health Research Premium Collection
Health Research Premium Collection (Alumni)
ProQuest Central Student
SciTech Premium Collection
ProQuest Health & Medical Complete (Alumni)
ProQuest Biological Science Collection
Health & Medical Collection (Alumni Edition)
PML(ProQuest Medical Library)
Science Database
Biological Science Database
ProQuest Central Premium
ProQuest One Academic (New)
Publicly Available Content Database
ProQuest Health & Medical Research Collection
ProQuest One Academic Middle East (New)
ProQuest One Health & Nursing
ProQuest One Academic Eastern Edition (DO NOT USE)
ProQuest One Applied & Life Sciences
ProQuest One Academic (retired)
ProQuest One Academic UKI Edition
ProQuest Central China
ProQuest Central Basic
MEDLINE - Academic
PubMed Central (Full Participant titles)
Directory of Open Access Journals
DatabaseTitle CrossRef
PubMed
Publicly Available Content Database
ProQuest Central Student
ProQuest One Academic Middle East (New)
ProQuest Central Essentials
ProQuest Health & Medical Complete (Alumni)
ProQuest Central (Alumni Edition)
SciTech Premium Collection
ProQuest One Community College
ProQuest One Health & Nursing
ProQuest Natural Science Collection
ProQuest Central China
ProQuest Biology Journals (Alumni Edition)
ProQuest Central
ProQuest One Applied & Life Sciences
ProQuest One Sustainability
ProQuest Health & Medical Research Collection
Health Research Premium Collection
Health and Medicine Complete (Alumni Edition)
Natural Science Collection
ProQuest Central Korea
Health & Medical Research Collection
Biological Science Collection
ProQuest Central (New)
ProQuest Medical Library (Alumni)
ProQuest Science Journals (Alumni Edition)
ProQuest Biological Science Collection
ProQuest Central Basic
ProQuest Science Journals
ProQuest One Academic Eastern Edition
ProQuest Hospital Collection
Health Research Premium Collection (Alumni)
Biological Science Database
ProQuest SciTech Collection
ProQuest Hospital Collection (Alumni)
ProQuest Health & Medical Complete
ProQuest Medical Library
ProQuest One Academic UKI Edition
ProQuest One Academic
ProQuest One Academic (New)
ProQuest Central (Alumni)
MEDLINE - Academic
DatabaseTitleList

Publicly Available Content Database

MEDLINE - Academic
PubMed
Database_xml – sequence: 1
  dbid: DOA
  name: DOAJ Open Access Full Text
  url: https://www.doaj.org/
  sourceTypes: Open Website
– sequence: 2
  dbid: NPM
  name: PubMed
  url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed
  sourceTypes: Index Database
– sequence: 3
  dbid: PIMPY
  name: Publicly Available Content Database
  url: http://search.proquest.com/publiccontent
  sourceTypes: Aggregation Database
DeliveryMethod fulltext_linktorsrc
Discipline Biology
EISSN 2045-2322
EndPage 19
ExternalDocumentID oai_doaj_org_article_e15c8de045f241838ea9385af1702c00
PMC12241452
40634505
10_1038_s41598_025_10348_9
Genre Journal Article
GroupedDBID 0R~
4.4
53G
5VS
7X7
88E
88I
8FE
8FH
8FI
8FJ
AAFWJ
AAJSJ
AAKDD
AASML
ABDBF
ABUWG
ACGFS
ACUHS
ADBBV
ADRAZ
AENEX
AEUYN
AFKRA
AFPKN
ALIPV
ALMA_UNASSIGNED_HOLDINGS
AOIJS
AZQEC
BAWUL
BBNVY
BCNDV
BENPR
BHPHI
BPHCQ
BVXVI
C6C
CCPQU
DIK
DWQXO
EBD
EBLON
EBS
ESX
FYUFA
GNUQQ
GROUPED_DOAJ
GX1
HCIFZ
HH5
HMCUK
HYE
KQ8
LK8
M1P
M2P
M7P
M~E
NAO
OK1
PHGZM
PHGZT
PIMPY
PPXIY
PQGLB
PQQKQ
PROAC
PSQYO
RNT
RNTTT
RPM
SNYQT
UKHRP
AAYXX
AFFHD
CITATION
PJZUB
NPM
3V.
7XB
88A
8FK
K9.
M48
PKEHL
PQEST
PQUKI
PRINS
Q9U
7X8
PUEGO
5PM
ID FETCH-LOGICAL-c543t-7970d0a1d943501d1b3f85d08eeebada7ffabbe09e2cd37f78feb2327351d66d3
IEDL.DBID 7X7
ISICitedReferencesCount 2
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001526484100014&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 2045-2322
IngestDate Fri Oct 03 12:42:52 EDT 2025
Tue Nov 04 02:04:28 EST 2025
Fri Sep 05 15:41:44 EDT 2025
Tue Oct 07 09:20:21 EDT 2025
Tue Jul 15 01:30:40 EDT 2025
Tue Nov 18 22:18:33 EST 2025
Sat Nov 29 07:34:25 EST 2025
Thu Jul 10 08:12:21 EDT 2025
IsDoiOpenAccess true
IsOpenAccess true
IsPeerReviewed true
IsScholarly true
Issue 1
Keywords Separation index
Encoder-Decoder architectures
Semantic segmentation
Model compression
Language English
License 2025. The Author(s).
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c543t-7970d0a1d943501d1b3f85d08eeebada7ffabbe09e2cd37f78feb2327351d66d3
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
OpenAccessLink https://www.proquest.com/docview/3228610895?pq-origsite=%requestingapplication%
PMID 40634505
PQID 3228610895
PQPubID 2041939
PageCount 19
ParticipantIDs doaj_primary_oai_doaj_org_article_e15c8de045f241838ea9385af1702c00
pubmedcentral_primary_oai_pubmedcentral_nih_gov_12241452
proquest_miscellaneous_3228823907
proquest_journals_3228610895
pubmed_primary_40634505
crossref_citationtrail_10_1038_s41598_025_10348_9
crossref_primary_10_1038_s41598_025_10348_9
springer_journals_10_1038_s41598_025_10348_9
PublicationCentury 2000
PublicationDate 2025-07-09
PublicationDateYYYYMMDD 2025-07-09
PublicationDate_xml – month: 07
  year: 2025
  text: 2025-07-09
  day: 09
PublicationDecade 2020
PublicationPlace London
PublicationPlace_xml – name: London
– name: England
PublicationTitle Scientific reports
PublicationTitleAbbrev Sci Rep
PublicationTitleAlternate Sci Rep
PublicationYear 2025
Publisher Nature Publishing Group UK
Nature Publishing Group
Nature Portfolio
Publisher_xml – name: Nature Publishing Group UK
– name: Nature Publishing Group
– name: Nature Portfolio
References R Reed (10348_CR20) 1993; 4
M Arsalan (10348_CR47) 2024; 133
10348_CR39
Z Wang (10348_CR43) 2004; 13
LC Chen (10348_CR7) 2017; 40
H Akaike (10348_CR61) 1974; 19
10348_CR36
10348_CR35
10348_CR38
M Saffar (10348_CR49) 2024; 14
10348_CR37
10348_CR32
10348_CR34
H Kim (10348_CR6) 2020; 9
10348_CR30
VN Vapnik (10348_CR59) 1999; 10
A Karimi (10348_CR53) 2024; 14
A Haider (10348_CR45) 2024; 167
10348_CR29
10348_CR25
10348_CR24
10348_CR27
10348_CR21
Y Zhou (10348_CR26) 2019; 51
J Chang (10348_CR31) 2022; 52
10348_CR23
10348_CR22
10348_CR60
10348_CR62
JC Caicedo (10348_CR56) 2019; 16
M Saffar (10348_CR48) 2023; 224
10348_CR8
10348_CR9
10348_CR18
10348_CR17
10348_CR58
10348_CR13
10348_CR57
10348_CR16
10348_CR15
10348_CR10
10348_CR12
10348_CR11
10348_CR55
10348_CR50
10348_CR52
10348_CR51
SS Sawant (10348_CR28) 2022; 52
H Sultan (10348_CR44) 2023; 35
G Shomron (10348_CR14) 2021; 34
V Badrinarayanan (10348_CR2) 2017; 39
A Parashar (10348_CR33) 2017; 45
GJ Brostow (10348_CR54) 2009; 30
10348_CR3
10348_CR4
10348_CR5
10348_CR46
10348_CR1
S Borkovkina (10348_CR19) 2020; 11
10348_CR42
10348_CR41
10348_CR40
References_xml – ident: 10348_CR52
  doi: 10.2139/ssrn.4796336
– ident: 10348_CR4
  doi: 10.1109/ICDCECE60827.2024.10548323
– ident: 10348_CR55
– ident: 10348_CR60
  doi: 10.1007/978-0-387-84858-7
– volume: 167
  start-page: 112300
  year: 2024
  ident: 10348_CR45
  publication-title: Appl. Soft Comput.
  doi: 10.1016/j.asoc.2024.112300
– ident: 10348_CR58
  doi: 10.1109/CVPR.2019.00982
– ident: 10348_CR17
– ident: 10348_CR51
  doi: 10.1109/ICSPIS56952.2022.10043890
– volume: 16
  start-page: 1247
  issue: 12
  year: 2019
  ident: 10348_CR56
  publication-title: Nat. Methods
  doi: 10.1038/s41592-019-0612-7
– ident: 10348_CR5
  doi: 10.1109/CVPRW50498.2020.00190
– ident: 10348_CR41
  doi: 10.1109/ICCV.2019.00038
– ident: 10348_CR32
  doi: 10.1109/BDCloud.2018.00110
– volume: 40
  start-page: 834
  issue: 4
  year: 2017
  ident: 10348_CR7
  publication-title: IEEE Trans. Pattern Anal. Mach. Intell.
  doi: 10.1109/TPAMI.2017.2699184
– ident: 10348_CR22
– volume: 133
  start-page: 108353
  year: 2024
  ident: 10348_CR47
  publication-title: Eng. Appl. Artif. Intell.
  doi: 10.1016/j.engappai.2024.108353
– ident: 10348_CR36
  doi: 10.4103/jmss.jmss_108_21
– volume: 224
  start-page: 119944
  year: 2023
  ident: 10348_CR48
  publication-title: Expert Syst. Appl.
  doi: 10.1016/j.eswa.2023.119944
– volume: 39
  start-page: 2481
  issue: 12
  year: 2017
  ident: 10348_CR2
  publication-title: IEEE Trans. Pattern Anal. Mach. Intell.
  doi: 10.1109/TPAMI.2016.2644615
– ident: 10348_CR10
– volume: 51
  start-page: 1626
  issue: 3
  year: 2019
  ident: 10348_CR26
  publication-title: IEEE Trans. Cybernetics
  doi: 10.1109/TCYB.2019.2928174
– ident: 10348_CR9
  doi: 10.3390/informatics8040077
– ident: 10348_CR39
– volume: 30
  start-page: 88
  issue: 2
  year: 2009
  ident: 10348_CR54
  publication-title: Pattern Recognit. Lett.
  doi: 10.1016/j.patrec.2008.04.005
– ident: 10348_CR18
– ident: 10348_CR37
  doi: 10.1016/j.neucom.2024.127817
– ident: 10348_CR1
  doi: 10.1007/978-3-319-24574-4_28
– ident: 10348_CR50
  doi: 10.1109/MVIP62238.2024.10491151
– ident: 10348_CR13
  doi: 10.1109/CVPR42600.2020.01318
– volume: 45
  start-page: 27
  issue: 2
  year: 2017
  ident: 10348_CR33
  publication-title: ACM SIGARCH Comput. Archit. News
  doi: 10.1145/3140659.3080254
– ident: 10348_CR11
  doi: 10.1016/j.neucom.2022.11.072
– ident: 10348_CR42
– ident: 10348_CR16
  doi: 10.1109/ISCAS48785.2022.9937283
– ident: 10348_CR57
– volume: 4
  start-page: 740
  issue: 5
  year: 1993
  ident: 10348_CR20
  publication-title: IEEE Trans. Neural Networks
  doi: 10.1109/72.248452
– ident: 10348_CR21
  doi: 10.1109/IJCNN.1991.170743
– volume: 9
  start-page: 8006
  year: 2020
  ident: 10348_CR6
  publication-title: IEEE Access.
  doi: 10.1109/ACCESS.2020.3048375
– ident: 10348_CR8
  doi: 10.3390/computers12030060
– ident: 10348_CR25
  doi: 10.1109/CVPR.2019.00293
– volume: 19
  start-page: 716
  issue: 6
  year: 1974
  ident: 10348_CR61
  publication-title: IEEE Trans. Autom. Control
  doi: 10.1109/TAC.1974.1100705
– ident: 10348_CR38
– ident: 10348_CR29
  doi: 10.1109/ICCV.2017.155
– volume: 13
  start-page: 600
  issue: 4
  year: 2004
  ident: 10348_CR43
  publication-title: IEEE Trans. Image Process.
  doi: 10.1109/TIP.2003.819861
– ident: 10348_CR34
– volume: 11
  start-page: 3968
  issue: 7
  year: 2020
  ident: 10348_CR19
  publication-title: Biomedical Opt. Express
  doi: 10.1364/BOE.395279
– ident: 10348_CR12
  doi: 10.1007/978-3-030-76423-4_12
– ident: 10348_CR62
– volume: 52
  start-page: 17751
  issue: 15
  year: 2022
  ident: 10348_CR31
  publication-title: Appl. Intell.
  doi: 10.1007/s10489-022-03508-1
– ident: 10348_CR24
– ident: 10348_CR46
  doi: 10.1109/CVPRW50498.2020.00088
– volume: 10
  start-page: 988
  issue: 5
  year: 1999
  ident: 10348_CR59
  publication-title: IEEE Trans. Neural Networks
  doi: 10.1109/72.788640
– ident: 10348_CR15
  doi: 10.1007/978-3-030-33642-4_13
– ident: 10348_CR35
  doi: 10.1109/CVPR.2018.00810
– ident: 10348_CR27
  doi: 10.1109/ICCV.2017.541
– ident: 10348_CR3
  doi: 10.53941/ijndi.2024.100009
– ident: 10348_CR30
  doi: 10.1145/3234804.3234824
– volume: 14
  start-page: 17590
  issue: 1
  year: 2024
  ident: 10348_CR49
  publication-title: Sci. Rep.
  doi: 10.1038/s41598-024-68172-6
– volume: 52
  start-page: 17557
  issue: 15
  year: 2022
  ident: 10348_CR28
  publication-title: Appl. Intell.
  doi: 10.1007/s10489-022-03229-5
– volume: 34
  start-page: 17737
  year: 2021
  ident: 10348_CR14
  publication-title: Adv. Neural. Inf. Process. Syst.
– volume: 14
  start-page: 8576
  issue: 1
  year: 2024
  ident: 10348_CR53
  publication-title: Sci. Rep.
  doi: 10.1038/s41598-024-59176-3
– volume: 35
  start-page: 101560
  issue: 5
  year: 2023
  ident: 10348_CR44
  publication-title: J. King Saud University-Computer Inform. Sci.
  doi: 10.1016/j.jksuci.2023.101560
– ident: 10348_CR23
– ident: 10348_CR40
SSID ssj0000529419
Score 2.4652107
Snippet We present a novel approach to compressing encoder–decoder architectures, particularly in semantic segmentation tasks, by leveraging the Separation Index...
We present a novel approach to compressing encoder-decoder architectures, particularly in semantic segmentation tasks, by leveraging the Separation Index...
Abstract We present a novel approach to compressing encoder–decoder architectures, particularly in semantic segmentation tasks, by leveraging the Separation...
SourceID doaj
pubmedcentral
proquest
pubmed
crossref
springer
SourceType Open Website
Open Access Repository
Aggregation Database
Index Database
Enrichment Source
Publisher
StartPage 24639
SubjectTerms 639/166/985
639/166/987
Accuracy
Compression
Encoder-Decoder architectures
Humanities and Social Sciences
Image processing
Methods
Model compression
multidisciplinary
Neural networks
Real time
Remote sensing
Science
Science (multidisciplinary)
Semantic segmentation
Semantics
Separation index
Tumors
SummonAdditionalLinks – databaseName: Directory of Open Access Journals
  dbid: DOA
  link: http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV1La9wwEB5KSKGX0nfdJEWF3loT2ZLXo2MTEnoooYcWchO2NEoXGm9Ybwr99xlJ3m22z0tPNpJshnl4Rp7RNwCv2SN7p5QpO1Ky1H1dlT16LPm72BlNhC7jzH5oz87w_Nx8vNXqK9aEZXjgzLhDqhqHnjjyCOxsUCF1RmHThaqVtZNpty5bc2szlVG9a6MrM52SkQoPR_ZU8TRZHYE3lWYr3_JECbD_d1Hmr8WSP2VMkyM6fQD3pwhSvMuUP4Q7NDyCu7mn5PfH4E8SKAS_RsRy8VzmOohFEBGy0tOy9JSuIjXBGQVHrWKkS-bw3PHNxeV0GmkQsSb-QnCEyMMZIpwHE7ziE_h8evLp-H05tVIoXaPVqmxNK73sKm90zCT6qlcBGy-RiPrOd20IXd-TNFQ7r9rQYuAtt-LYpqn8bObVU9gZFgM9B6GCVxhmLAuqdE_aOBMC9kbq3hsZfAHVmq3WTTjjsd3FV5vy3QptFoVlUdgkCmsKeLN55iqjbPx19VGU1mZlRMhOA6w3dtIb-y-9KWB_LWs7me1o-euGrL1omgJebabZ4GIWpRtocZ3XYK2MbAt4llVjQwlHR0pzTFkAbinNFqnbM8P8SwL1ThlO3dQFvF3r1w-6_syLF_-DF3twr46GEf9am33YWS2v6QB23bfVfFy-TJZ1AzgCJpA
  priority: 102
  providerName: Directory of Open Access Journals
Title Efficient compression of encoder-decoder models for semantic segmentation using the separation index
URI https://link.springer.com/article/10.1038/s41598-025-10348-9
https://www.ncbi.nlm.nih.gov/pubmed/40634505
https://www.proquest.com/docview/3228610895
https://www.proquest.com/docview/3228823907
https://pubmed.ncbi.nlm.nih.gov/PMC12241452
https://doaj.org/article/e15c8de045f241838ea9385af1702c00
Volume 15
WOSCitedRecordID wos001526484100014&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVAON
  databaseName: DOAJ Open Access Full Text
  customDbUrl:
  eissn: 2045-2322
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0000529419
  issn: 2045-2322
  databaseCode: DOA
  dateStart: 20110101
  isFulltext: true
  titleUrlDefault: https://www.doaj.org/
  providerName: Directory of Open Access Journals
– providerCode: PRVHPJ
  databaseName: ROAD: Directory of Open Access Scholarly Resources
  customDbUrl:
  eissn: 2045-2322
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0000529419
  issn: 2045-2322
  databaseCode: M~E
  dateStart: 20110101
  isFulltext: true
  titleUrlDefault: https://road.issn.org
  providerName: ISSN International Centre
– providerCode: PRVPQU
  databaseName: Biological Science Database
  customDbUrl:
  eissn: 2045-2322
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0000529419
  issn: 2045-2322
  databaseCode: M7P
  dateStart: 20110101
  isFulltext: true
  titleUrlDefault: http://search.proquest.com/biologicalscijournals
  providerName: ProQuest
– providerCode: PRVPQU
  databaseName: Health & Medical Collection
  customDbUrl:
  eissn: 2045-2322
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0000529419
  issn: 2045-2322
  databaseCode: 7X7
  dateStart: 20110101
  isFulltext: true
  titleUrlDefault: https://search.proquest.com/healthcomplete
  providerName: ProQuest
– providerCode: PRVPQU
  databaseName: ProQuest Central
  customDbUrl:
  eissn: 2045-2322
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0000529419
  issn: 2045-2322
  databaseCode: BENPR
  dateStart: 20110101
  isFulltext: true
  titleUrlDefault: https://www.proquest.com/central
  providerName: ProQuest
– providerCode: PRVPQU
  databaseName: Publicly Available Content Database
  customDbUrl:
  eissn: 2045-2322
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0000529419
  issn: 2045-2322
  databaseCode: PIMPY
  dateStart: 20110101
  isFulltext: true
  titleUrlDefault: http://search.proquest.com/publiccontent
  providerName: ProQuest
– providerCode: PRVPQU
  databaseName: Science Database
  customDbUrl:
  eissn: 2045-2322
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0000529419
  issn: 2045-2322
  databaseCode: M2P
  dateStart: 20110101
  isFulltext: true
  titleUrlDefault: https://search.proquest.com/sciencejournals
  providerName: ProQuest
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV3db9MwED_BChIvfMMCowoSbxDNiZ36_IQY6gQSqyIEUnmKEn90lViyNR0S_z1nJ-1UPvbCi1M5buXk7uxf786_A3hFO7LRnKukspwlos7SpEaDCa2LlRLWou55Zj_J2Qznc1UMDrduSKvcrIlhoTat9j7yQ1I8pB9Glb89v0h81SgfXR1KaNyEkS-b7fVczuXWx-KjWCJVw1kZxvGwo_3KnynLPP0mF2TrO_tRoO3_G9b8M2Xyt7hp2I6O7_3vg9yHuwMQjd_1mvMAbtjmIdzuS1P-fARmGrglaB6xzzrvs2WbuHWxZ740dpUYG65xqKXTxQR-486ekaCWmj4szoZDTU3sU-sXMQFN6u6ZxqkzsDQ-hq_H0y_vPyRDRYZE54KvE6kkM6xKjRI-IGnSmjvMDUNrbV2ZSjpX1bVlymbacOkkOvrnzgki5amZTAx_AntN29h9iLkzHN1EpGhTUVuhtHIOa8VEbRRzJoJ0I5dSD3TlvmrG9zKEzTmWvSxLkmUZZFmqCF5vv3Pek3VcO_rIi3s70hNth452tSgHuy1tmms0loCvI6yDHG2lOOaVSyXLNGMRHGykXA7W35VXIo7g5fY22a0PxlSNbS_7MZhxxWQET3vd2s6EQBYXBE0jwB2t25nq7p1meRq4wUOgVORZBG82Cno1r3-_i2fXP8ZzuJN5m_FubXUAe-vVpX0Bt_SP9bJbjYPRhRbHMDqazorP4-DboPYkK3wrqR0VH0-Kb78AhfU8Vw
linkProvider ProQuest
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMw1V1Lb9QwEB6VAoILb0qggJHgBFGd2FnbB4R4tGrVZcWhSHsziR_LSm22bLag_il-I2Mn2Wp59NYDp1053sjOfvOIZ-YbgOdoka1hTKWlYzTlVZ6llbQyRb1YKu6cNC3P7FCMRnI8Vp_W4GdfCxPSKnudGBW1nZlwRr6FwJN4Y6mKN8ff0tA1KkRX-xYaLSz23ekPfGVrXu99wP_3RZ7vbB-83027rgKpKThbpEIJammZWcVDUM1mFfOysFQ656rSlsL7sqocVS43lgkvpMe3T4ZmvsjsYGAZ3vcSXEY9LkLHBDEWyzOdEDXjmepqcyiTWw3ax1DDlge6T8ZRt6zYv9gm4G--7Z8pmr_FaaP527n5vz24W3Cjc7TJ21YybsOaq-_A1bb15uldsNuROwP3TUJWfZsNXJOZJ4HZ07p5al38JLFXUEPQuSeNO0IgTg1-mRx1RVs1CaUDE4KONA63TOo4GFko78HnC9nifVivZ7V7AIR5y6Qf8Ey6jFeOK6O8l5WivLKKeptA1uNAm46OPXQFOdQxLYBJ3WJHI3Z0xI5WCbxc_ua4JSM5d_a7AK_lzEAkHgdm84nu9JJ2WWGkdejYe_TlJJOuVEwWpc8EzQ2lCWz2qNKddmv0GaQSeLa8jHopBJvK2s1O2jkyZ4qKBDZaLC9Xgk4k4-h6JyBXUL6y1NUr9fRr5D6PgWBe5Am86gXibF3_fhYPz9_GU7i2e_BxqId7o_1HcD0P8hqO8NUmrC_mJ-4xXDHfF9Nm_iQKPIEvFy0ovwC9uZSt
linkToPdf http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMw1V1Jb9QwFH4qZREX9iVQwEhwotE4sTOxDwgB7Yiq1agHkOZmEi_DSDQpkymof41fx7OzVMPSWw-cZuR4IsfzvSV-730P4AVaZKMZk3FhGY15mSZxKYyIUS8WklsrdMsze5BPp2I2k4cb8LOvhfFplb1ODIra1NqfkY8QeAJvLGQ2cl1axOHO5M3xt9h3kPKR1r6dRguRfXv6A1_fmtd7O_hfv0zTye7H9x_irsNArDPOVnEuc2pokRjJfYDNJCVzIjNUWGvLwhS5c0VZWiptqg3LXS4cvokyNPlZYsZjw_C-l-CypxT0MpbP8uF8x0fQeCK7Oh3KxKhBW-nr2VJP_ck46pk1WxhaBvzNz_0zXfO3mG0whZOb__Mm3oIbnQNO3rYScxs2bHUHrrYtOU_vgtkNnBq4B8Rn27dZwhWpHfGMn8YuY2PDJwk9hBqCTj9p7BECdKHxy_yoK-aqiC8pmBN0sHG4ZVjHwcBOeQ8-Xcgj3ofNqq7sQyDMGSbcmCfCJry0XGrpnCgl5aWR1JkIkh4TSnc07b5byFcV0gWYUC2OFOJIBRwpGcGr4TfHLUnJubPfeagNMz3BeBiol3PV6Stlk0wLY9Hhd-jjCSZsIZnICpfkNNWURrDVI0x1Wq9RZ_CK4PlwGfWVD0IVla1P2jkiZZLmETxocT2sBJ1LxtElj0CsIX5tqetXqsWXwIkeAsQ8SyPY7oXjbF3_3otH5z_GM7iG8qEO9qb7j-F66kXXn-zLLdhcLU_sE7iiv68WzfJpkH0Cny9aTn4BLwqdaQ
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Efficient+compression+of+encoder-decoder+models+for+semantic+segmentation+using+the+separation+index&rft.jtitle=Scientific+reports&rft.au=Jamshidi%2C+Movahed&rft.au=Kalhor%2C+Ahmad&rft.au=Vahabie%2C+Abdol-Hossein&rft.date=2025-07-09&rft.eissn=2045-2322&rft.volume=15&rft.issue=1&rft.spage=24639&rft_id=info:doi/10.1038%2Fs41598-025-10348-9&rft_id=info%3Apmid%2F40634505&rft.externalDocID=40634505
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=2045-2322&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=2045-2322&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=2045-2322&client=summon