Potential of SEM and Deep Learning in Archaeobotanical Identification of Ancient Wheat Varieties

dc.authorscopusid 55293387500
dc.authorscopusid 56247318100
dc.authorscopusid 55537456900
dc.authorscopusid 8688503700
dc.authorscopusid 57112228600
dc.authorscopusid 56288729800
dc.authorscopusid 57203178520
dc.contributor.author Anagun, Yildiray
dc.contributor.author Isik, Sahin
dc.contributor.author Olgun, Murat
dc.contributor.author Ulker, Mehmet
dc.contributor.author Koyuncu, Onur
dc.contributor.author Dikmen, Gokhan
dc.contributor.author Biber, Hanifi
dc.date.accessioned 2025-12-30T16:05:33Z
dc.date.available 2025-12-30T16:05:33Z
dc.date.issued 2025
dc.department T.C. Van Yüzüncü Yıl Üniversitesi en_US
dc.department-temp [Anagun, Yildiray; Isik, Sahin] Eskisehir Osmangazi Univ, Dept Comp Engn, TR-26040 Eskisehir, Turkiye; [Olgun, Murat] Eskisehir Osmangazi Univ, Dept Field Crops, Eskisehir, Turkiye; [Ulker, Mehmet; Ozdemir, Burak; Salih, Sana Jamal; Oral, Erol] Yuzuncu Yil Univ, Fac Agr, Dept Field Crops, Van, Turkiye; [Koyuncu, Onur] Eskisehir Osmangazi Univ, Dept Bot, Eskisehir, Turkiye; [Dikmen, Gokhan] Eskisehir Osmangazi Univ, Cent Res Lab Applicat & Res Ctr ARUM, Eskisehir, Turkiye; [Cavusoglu, Rafet; Biber, Hanifi] Yuzuncu Yil Univ, Dept Archeol, Van, Turkiye; [Altuner, Fevzi] Yuzuncu Yil Univ, Gevas Vocat Sch, Dept Plant & Anim Prod, Van, Turkiye en_US
dc.description.abstract This study investigated the morphological similarity between bread wheat landraces from the Van Lake Basin and ancient Urartian wheat seeds (9th century BCE) discovered at & Ccedil;avu & scedil;tepe Fortress, utilising a Convolutional Neural Network (CNN)-based framework. Scanning Electron Microscopy (SEM) datasets were created using 15 lines from 10 landraces and the ancient seeds; EfficientNetB0, ResNet18 and InceptionResNetV2 models were employed to extract discriminative surface texture features. The ancient wheat samples showed the highest surface-texture similarity to the Muradiye-1-1 line (Red Kirik wheat) across the tested models (39.8% to 44.1%). These results suggest a phenotypic convergence between ancient and modern landraces under similar agroecological conditions, demonstrating the utility of CNN models for archaeobotanical analysis. en_US
dc.description.sponsorship Van Yuzuncu Yil University Scientific Research Projects Directorate (Van YYU-SRPD) [FBA 2019-8276] en_US
dc.description.sponsorship In this study, ancient wheat seeds were used with official permission from the Van Governorship of the Republic of Turkey, the Van Provincial Directorate of Culture and Tourism, and the Van Museum Directorate. Therefore, the dataset is not publicly available. The wheat lines examined in the study were selected from landraces cultivated in the Van Lake Basin, as part of the project titled 'Collection, Identifi-cation, and Preservation of Bread Wheat Landraces Grown in the Van Lake Basin, and Analysis of the Relationships Between Local Varieties and Soil Characteristics.' This project was supported by Van Yuzuncu Yil University Scientific Research Projects Directorate (Van YYU-SRPD, Project No: FBA 2019-8276). The training and validation datasets were prepared and analysed using the scanning electron microscope (SEM) at the Central Research Laboratory Application and Research Center of Eskisehir Osmangazi University. en_US
dc.description.woscitationindex Science Citation Index Expanded - Arts & Humanities Citation Index
dc.identifier.doi 10.1080/14614103.2025.2600115
dc.identifier.issn 1461-4103
dc.identifier.issn 1749-6314
dc.identifier.scopus 2-s2.0-105025032057
dc.identifier.scopusquality Q2
dc.identifier.uri https://doi.org/10.1080/14614103.2025.2600115
dc.identifier.uri https://hdl.handle.net/20.500.14720/29323
dc.identifier.wos WOS:001640208300001
dc.identifier.wosquality Q4
dc.language.iso en en_US
dc.publisher Routledge Journals, Taylor & Francis Ltd en_US
dc.relation.ispartof Environmental Archaeology en_US
dc.relation.publicationcategory Makale - Uluslararası Hakemli Dergi - Kurum Öğretim Elemanı en_US
dc.rights info:eu-repo/semantics/closedAccess en_US
dc.subject Ancient Wheat en_US
dc.subject Archaeobotany en_US
dc.subject Wheat Landraces en_US
dc.subject Convolutional Neural Network en_US
dc.subject Scanning Electron Microscopy en_US
dc.subject Surface Texture Classification en_US
dc.title Potential of SEM and Deep Learning in Archaeobotanical Identification of Ancient Wheat Varieties en_US
dc.type Article en_US
dspace.entity.type Publication
gdc.coar.access metadata only access
gdc.coar.type text::journal::journal article

Files