Academic Journal
Προβολή σε EDS
Συνδεδεμένο Πλήρες Κείμενο

Deep learning based instance segmentation of mandarin fruit slices for precision assessment and morphological quantification.

Λεπτομέρειες βιβλιογραφικής εγγραφής
Τίτλος: Deep learning based instance segmentation of mandarin fruit slices for precision assessment and morphological quantification.
Συγγραφείς: Le AT; Faculty of Biology and Biotechnology, University of Science, Ho Chi Minh City, 700000, Vietnam.; Vietnam National University, Ho Chi Minh City, 700000, Vietnam., Ahn J; Department of Management Information Systems, Jeju National University, Jeju, 63243, Republic of Korea. jha@jejunu.ac.kr., Thai TT; Vietnam National University, Ho Chi Minh City, 700000, Vietnam. thaithanhtuan@jejunu.ac.kr.; Multimedia Communications Laboratory, University of Information Technology, Ho Chi Minh City, 700000, Vietnam. thaithanhtuan@jejunu.ac.kr.; Department of Plant Resources and Environment, Jeju National University, Jeju, 63243, Republic of Korea. thaithanhtuan@jejunu.ac.kr.
Πηγή: Scientific reports [Sci Rep] 2026 Apr 17; Vol. 16 (1). Date of Electronic Publication: 2026 Apr 17.
Τύπος έκδοσης: Journal Article
Γλώσσα: English
Στοιχεία περιοδικού: Publisher: Nature Publishing Group Country of Publication: England NLM ID: 101563288 Publication Model: Electronic Cited Medium: Internet ISSN: 2045-2322 (Electronic) Linking ISSN: 20452322 NLM ISO Abbreviation: Sci Rep Subsets: MEDLINE
Imprint Name(s): Original Publication: London : Nature Publishing Group, copyright 2011-
Ιατρικοί όροι (MeSH): Citrus*/anatomy & histology , Fruit*/anatomy & histology , Image Processing, Computer-Assisted*/methods , Deep Learning* , Detection Algorithms*
Περίληψη: Accurate instance segmentation of mandarin fruit slices is essential for quantifying segment morphology and central core structure, which are key traits in cultivar evaluation, fruit quality assessment, and postharvest application. Manual measurement of these anatomical features, however, is time-consuming and prone to inconsistency. In this study, we developed an automated segmentation framework based on YOLOv8 to detect and semantic segment the fruit segments and the central core in high-resolution mandarin transversely cut images. The model was trained on a curated datasetderived from 58 original high-resolution cross-sectional images (5100 × 7019 pixels), which were systematically partitioned into 280 cropped sub-images (1712 × 1778 pixels), each containing a single complete citrus slice, and demonstrated excellent performance. YOLOv8 achieved near-perfect detection metrics, with bounding box metrics precision ~ 0.997, recall ~ 1.00, mAP50 ~ 0.995, and mAP50-95 ~ 0.922. Semantic segmentation accuracy was similarly strong, with precision = 0.997, Recall ~ 1.00, mAP50 ~ 0.995, and mAP50-95 = 0.965. Training and validation losses converged steadily, indicating stable learning without overfitting. The nonsignificant differences between YOLOv8-predicted measurements and ground truth data, together with low mean absolute error (MAE) values, demonstrate that the model not only performs well in semantic segmentation metrics but also maintains high accuracy in quantitative measurements, which is critical for cultivar discrimination and genetic studies. Our work provides a robust, high-precision, and reproducible framework for mandarin fruit slice phenotyping, offering significant potential for applications in agricultural research, breeding programs, and automated fruit quality evaluation.
(© 2026. The Author(s).)
Competing Interests: Declarations. Competing interests: The authors declare no competing interests. Ethical approval and consent to participate: Not applicable. Consent for publication: All authors have reviewed the final manuscript and consent to its publication. No individual person’s data are included that would require additional consent.
References: Front Plant Sci. 2022 Jun 09;13:765523. (PMID: 35755692)
Front Plant Sci. 2022 Aug 10;13:952942. (PMID: 36035725)
Front Plant Sci. 2024 Jun 05;15:1397816. (PMID: 38903428)
Trends Genet. 2001 Sep;17(9):536-40. (PMID: 11525837)
Crit Rev Anal Chem. 2023;53(7):1489-1514. (PMID: 35157545)
Plant Phenomics. 2023 Jun 07;5:0057. (PMID: 37292188)
J Sci Food Agric. 2018 Jan;98(1):18-26. (PMID: 28631804)
Front Plant Sci. 2024 Jun 05;15:1411178. (PMID: 38903423)
Sci Rep. 2025 Apr 12;15(1):12659. (PMID: 40221550)
Front Plant Sci. 2021 Feb 11;12:622062. (PMID: 33643351)
Front Plant Sci. 2019 Sep 26;10:1167. (PMID: 31611894)
BMC Genet. 2015 Apr 18;16:39. (PMID: 25902849)
Foods. 2022 Jun 22;11(13):. (PMID: 35804656)
PLoS One. 2015 Mar 04;10(3):e0118432. (PMID: 25738806)
Compr Rev Food Sci Food Saf. 2017 Nov;16(6):1345-1358. (PMID: 33371593)
PLoS One. 2024 Aug 26;19(8):e0308826. (PMID: 39186505)
Grant Information: RS-2024-00348897 National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT)
Contributed Indexing: Keywords: Instance segmentation; Mandarin fruit; Phenotyping; YOLOv8
Entry Date(s): Date Created: 20260417 Date Completed: 20260612 Latest Revision: 20260813
Update Code: 20260813
PubMed Central ID: PMC13246784
DOI: 10.1038/s41598-026-46784-4
PMID: 41998006
Βάση Δεδομένων: MEDLINE
FullText Links:
  – Type: other
    Url: https://resolver.ebsco.com:443/public/rma-ftfapi/ejs/direct?AccessToken=4EEEB37D962950660CD9&Show=Object
Text:
  Availability: 0
CustomLinks:
  – Url: https://dx.doi.org/doi:10.1038/s41598-026-46784-4
    Name: EDS - Springer Nature Journals (s7799221)
    Category: fullText
    Text: View record at Springer
Header DbId: cmedm
DbLabel: MEDLINE
An: 41998006
AccessLevel: 3
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Deep learning based instance segmentation of mandarin fruit slices for precision assessment and morphological quantification.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AU" term="%22Le+AT%22">Le AT</searchLink>; Faculty of Biology and Biotechnology, University of Science, Ho Chi Minh City, 700000, Vietnam.; Vietnam National University, Ho Chi Minh City, 700000, Vietnam.<br /><searchLink fieldCode="AU" term="%22Ahn+J%22">Ahn J</searchLink>; Department of Management Information Systems, Jeju National University, Jeju, 63243, Republic of Korea. jha@jejunu.ac.kr.<br /><searchLink fieldCode="AU" term="%22Thai+TT%22">Thai TT</searchLink>; Vietnam National University, Ho Chi Minh City, 700000, Vietnam. thaithanhtuan@jejunu.ac.kr.; Multimedia Communications Laboratory, University of Information Technology, Ho Chi Minh City, 700000, Vietnam. thaithanhtuan@jejunu.ac.kr.; Department of Plant Resources and Environment, Jeju National University, Jeju, 63243, Republic of Korea. thaithanhtuan@jejunu.ac.kr.
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22101563288%22">Scientific reports</searchLink> [Sci Rep] 2026 Apr 17; Vol. 16 (1). <i>Date of Electronic Publication: </i>2026 Apr 17.
– Name: TypePub
  Label: Publication Type
  Group: TypPub
  Data: Journal Article
– Name: Language
  Label: Language
  Group: Lang
  Data: English
– Name: TitleSource
  Label: Journal Info
  Group: Src
  Data: <i>Publisher: </i><searchLink fieldCode="PB" term="%22Nature+Publishing+Group%22">Nature Publishing Group </searchLink><i>Country of Publication: </i>England <i>NLM ID: </i>101563288 <i>Publication Model: </i>Electronic <i>Cited Medium: </i>Internet <i>ISSN: </i>2045-2322 (Electronic) <i>Linking ISSN: </i><searchLink fieldCode="IS" term="%2220452322%22">20452322 </searchLink><i>NLM ISO Abbreviation: </i>Sci Rep <i>Subsets: </i>MEDLINE
– Name: PublisherInfo
  Label: Imprint Name(s)
  Group: PubInfo
  Data: <i>Original Publication</i>: London : Nature Publishing Group, copyright 2011-
– Name: SubjectMESH
  Label: MeSH Terms
  Group: Su
  Data: <searchLink fieldCode="MM" term="%22Citrus%22">Citrus*</searchLink>/<searchLink fieldCode="MM" term="%22Citrus+anatomy+%26+histology%22">anatomy & histology</searchLink> <br /><searchLink fieldCode="MM" term="%22Fruit%22">Fruit*</searchLink>/<searchLink fieldCode="MM" term="%22Fruit+anatomy+%26+histology%22">anatomy & histology</searchLink> <br /><searchLink fieldCode="MM" term="%22Image+Processing%2C+Computer-Assisted%22">Image Processing, Computer-Assisted*</searchLink>/<searchLink fieldCode="MM" term="%22Image+Processing%2C+Computer-Assisted+methods%22">methods</searchLink> <br /><searchLink fieldCode="MM" term="%22Deep+Learning%22">Deep Learning*</searchLink> <br /><searchLink fieldCode="MM" term="%22Detection+Algorithms%22">Detection Algorithms*</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Accurate instance segmentation of mandarin fruit slices is essential for quantifying segment morphology and central core structure, which are key traits in cultivar evaluation, fruit quality assessment, and postharvest application. Manual measurement of these anatomical features, however, is time-consuming and prone to inconsistency. In this study, we developed an automated segmentation framework based on YOLOv8 to detect and semantic segment the fruit segments and the central core in high-resolution mandarin transversely cut images. The model was trained on a curated datasetderived from 58 original high-resolution cross-sectional images (5100 × 7019 pixels), which were systematically partitioned into 280 cropped sub-images (1712 × 1778 pixels), each containing a single complete citrus slice, and demonstrated excellent performance. YOLOv8 achieved near-perfect detection metrics, with bounding box metrics precision ~ 0.997, recall ~ 1.00, mAP50 ~ 0.995, and mAP50-95 ~ 0.922. Semantic segmentation accuracy was similarly strong, with precision = 0.997, Recall ~ 1.00, mAP50 ~ 0.995, and mAP50-95 = 0.965. Training and validation losses converged steadily, indicating stable learning without overfitting. The nonsignificant differences between YOLOv8-predicted measurements and ground truth data, together with low mean absolute error (MAE) values, demonstrate that the model not only performs well in semantic segmentation metrics but also maintains high accuracy in quantitative measurements, which is critical for cultivar discrimination and genetic studies. Our work provides a robust, high-precision, and reproducible framework for mandarin fruit slice phenotyping, offering significant potential for applications in agricultural research, breeding programs, and automated fruit quality evaluation.<br /> (© 2026. The Author(s).)
– Name: Abstract
  Label: Competing Interests
  Group: Ab
  Data: Declarations. Competing interests: The authors declare no competing interests. Ethical approval and consent to participate: Not applicable. Consent for publication: All authors have reviewed the final manuscript and consent to its publication. No individual person’s data are included that would require additional consent.
– Name: Ref
  Label: References
  Group: RefInfo
  Data: Front Plant Sci. 2022 Jun 09;13:765523. (PMID: <searchLink fieldCode="PM" term="%2235755692%22">35755692)</searchLink><br />Front Plant Sci. 2022 Aug 10;13:952942. (PMID: <searchLink fieldCode="PM" term="%2236035725%22">36035725)</searchLink><br />Front Plant Sci. 2024 Jun 05;15:1397816. (PMID: <searchLink fieldCode="PM" term="%2238903428%22">38903428)</searchLink><br />Trends Genet. 2001 Sep;17(9):536-40. (PMID: <searchLink fieldCode="PM" term="%2211525837%22">11525837)</searchLink><br />Crit Rev Anal Chem. 2023;53(7):1489-1514. (PMID: <searchLink fieldCode="PM" term="%2235157545%22">35157545)</searchLink><br />Plant Phenomics. 2023 Jun 07;5:0057. (PMID: <searchLink fieldCode="PM" term="%2237292188%22">37292188)</searchLink><br />J Sci Food Agric. 2018 Jan;98(1):18-26. (PMID: <searchLink fieldCode="PM" term="%2228631804%22">28631804)</searchLink><br />Front Plant Sci. 2024 Jun 05;15:1411178. (PMID: <searchLink fieldCode="PM" term="%2238903423%22">38903423)</searchLink><br />Sci Rep. 2025 Apr 12;15(1):12659. (PMID: <searchLink fieldCode="PM" term="%2240221550%22">40221550)</searchLink><br />Front Plant Sci. 2021 Feb 11;12:622062. (PMID: <searchLink fieldCode="PM" term="%2233643351%22">33643351)</searchLink><br />Front Plant Sci. 2019 Sep 26;10:1167. (PMID: <searchLink fieldCode="PM" term="%2231611894%22">31611894)</searchLink><br />BMC Genet. 2015 Apr 18;16:39. (PMID: <searchLink fieldCode="PM" term="%2225902849%22">25902849)</searchLink><br />Foods. 2022 Jun 22;11(13):. (PMID: <searchLink fieldCode="PM" term="%2235804656%22">35804656)</searchLink><br />PLoS One. 2015 Mar 04;10(3):e0118432. (PMID: <searchLink fieldCode="PM" term="%2225738806%22">25738806)</searchLink><br />Compr Rev Food Sci Food Saf. 2017 Nov;16(6):1345-1358. (PMID: <searchLink fieldCode="PM" term="%2233371593%22">33371593)</searchLink><br />PLoS One. 2024 Aug 26;19(8):e0308826. (PMID: <searchLink fieldCode="PM" term="%2239186505%22">39186505)</searchLink>
– Name: GrantInfo
  Label: Grant Information
  Group: Grant
  Data: RS-2024-00348897 National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT)
– Name: SubjectMinor
  Label: Contributed Indexing
  Group:
  Data: <i>Keywords: </i>Instance segmentation; Mandarin fruit; Phenotyping; YOLOv8
– Name: DateEntry
  Label: Entry Date(s)
  Group: Date
  Data: <i>Date Created: </i>20260417 <i>Date Completed: </i>20260612 <i>Latest Revision: </i>20260813
– Name: DateUpdate
  Label: Update Code
  Group: Date
  Data: 20260813
– Name: PubmedCentralID
  Label: PubMed Central ID
  Group: ID
  Data: PMC13246784
– Name: DOI
  Label: DOI
  Group: ID
  Data: 10.1038/s41598-026-46784-4
– Name: AN
  Label: PMID
  Group: ID
  Data: 41998006
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=cmedm&AN=41998006
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1038/s41598-026-46784-4
    Languages:
      – Code: eng
        Text: English
    Subjects:
      – SubjectFull: Citrus anatomy & histology
        Type: general
      – SubjectFull: Fruit anatomy & histology
        Type: general
      – SubjectFull: Image Processing, Computer-Assisted methods
        Type: general
      – SubjectFull: Deep Learning
        Type: general
      – SubjectFull: Detection Algorithms
        Type: general
    Titles:
      – TitleFull: Deep learning based instance segmentation of mandarin fruit slices for precision assessment and morphological quantification.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Le AT
      – PersonEntity:
          Name:
            NameFull: Ahn J
      – PersonEntity:
          Name:
            NameFull: Thai TT
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 17
              M: 04
              Text: 2026 Apr 17
              Type: published
              Y: 2026
          Identifiers:
            – Type: issn-electronic
              Value: 2045-2322
          Numbering:
            – Type: volume
              Value: 16
            – Type: issue
              Value: 1
          Titles:
            – TitleFull: Scientific reports
              Type: main
ResultId 1