{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,23]],"date-time":"2026-10-23T04:06:01Z","timestamp":1792728361479,"version":"4.5.0"},"reference-count":24,"publisher":"Korean Society of Animal Science and Technology","issue":"4","license":[{"start":{"date-parts":[[2026,7,1]],"date-time":"2026-07-01T00:00:00Z","timestamp":1782864000000},"content-version":"tdm","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by-nc\/4.0"}],"content-domain":{"domain":["www.ejast.org"],"crossmark-restriction":true},"short-container-title":["J Anim Sci Technol"],"published-print":{"date-parts":[[2026,7]]},"abstract":"<jats:p>In the poultry industry, accurate distinction between fertilized, unfertilized,\n                    and hatching failure eggs is a key factor for improving hatching efficiency and\n                    reducing resource waste. This study comprehensively evaluated 12 CNN\n                    architectures for non-destructive egg fertility classification based on\n                    candling-style images captured on days 5\u20136 of incubation, under four\n                    experimental scenarios: Case A (image-level random split, three classes), Case B\n                    (egg-level separation, three classes), Case C (image-level random split, two\n                    classes), and Case D (egg-level separation, two classes). In Cases A and B,\n                    overall accuracy ranged from 93.65%\u201398.65% and 90.82%\u201395.79%,\n                    respectively. Per-class recall for fertilized eggs exceeded 98% across all\n                    models, while hatching failure egg recall was highly variable\n                    (5.88%\u201376.47% in Case A; 0%\u201327.78% in Case B), reflecting severe\n                    class imbalance and visual ambiguity. In Cases C and D (fertilized vs.\n                    unfertilized only), overall accuracy improved to 97.02%\u201399.80% and\n                    95.84%\u201398.81%, respectively, with unfertilized egg recall ranging from\n                    74.47%\u201397.87% (Case C) and 58.33%\u201387.50% (Case D). Egg-level\n                    evaluation (Cases B and D) consistently yielded lower accuracy than image-level\n                    splitting (Cases A and C), confirming that image-level splitting overestimates\n                    real-world generalization performance. CNN selection should be guided by the\n                    balance between computational resources and required accuracy: lightweight\n                    models (SqueezeNet: 4.7 MB; ShuffleNet: 5.5 MB) suit resource-constrained\n                    deployments, while larger models (Inception-v3, VGG-16, VGG-19) offer superior\n                    accuracy for high-performance applications. Classification of hatching failure\n                    eggs remains challenging due to their visual similarity to both fertilized and\n                    unfertilized eggs depending on the timing of developmental arrest, and\n                    represents an important direction for future research.<\/jats:p>","DOI":"10.5187\/jast.2600100","type":"journal-article","created":{"date-parts":[[2026,6,2]],"date-time":"2026-06-02T01:28:12Z","timestamp":1780363692000},"page":"1272-1293","update-policy":"https:\/\/doi.org\/10.5187\/crossmark_policy","source":"Crossref","is-referenced-by-count":1,"title":["Egg fertility assessment system using deep learning\n                    technology"],"prefix":"10.5187","volume":"68","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-7966-0014","authenticated-orcid":false,"given":"Jin-Hyun","family":"Park","sequence":"first","affiliation":[{"id":[{"id":"https:\/\/ror.org\/00saywf64","id-type":"ROR","asserted-by":"publisher"}],"name":"School of Mechatronics Engineering, College of IT Engineering, Gyeongsang National University, Jinju, Korea"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6735-9761","authenticated-orcid":false,"given":"Sea Hwan","family":"Sohn","sequence":"additional","affiliation":[{"name":"HenTech, Jinju, Korea"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5182-1739","authenticated-orcid":false,"given":"Hee-Mun","family":"Park","sequence":"additional","affiliation":[{"name":"Department of Automation Machine Maintenance, Suncheon Campus of Korea Polytechnics, Suncheon, Korea"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9696-9638","authenticated-orcid":false,"given":"Sang-Hyon","family":"Oh","sequence":"additional","affiliation":[{"id":[{"id":"https:\/\/ror.org\/00saywf64","id-type":"ROR","asserted-by":"publisher"}],"name":"Division of Animal Science, College of Agriculture and Life Science, Gyeongsang National University, Jinju, Korea"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"3132","published-online":{"date-parts":[[2026,7,31]]},"reference":[{"key":"key2.0261002114004e+13_B1","doi-asserted-by":"crossref","first-page":"119","DOI":"10.1080\/00439339.2023.2163955","article-title":"The use of silkworm pupae (Bombyx mori) meal as an alternative\n                        protein source for poultry","volume":"79","author":"Banday","year":"2023","journal-title":"World\u2019s Poult Sci J."},{"key":"key2.0261002114004e+13_B2","doi-asserted-by":"crossref","first-page":"5","DOI":"10.1079\/WPS20020002","article-title":"Applications of biotechnology in the poultry\n                        industry","volume":"58","author":"Burt","year":"2002","journal-title":"World\u2019s Poult Sci J."},{"key":"key2.0261002114004e+13_B3","doi-asserted-by":"crossref","first-page":"483","DOI":"10.3923\/ijps.2011.483.492","article-title":"Review of the factors that influence egg fertility and\n                        hatchabilty in poultry","volume":"10","author":"Ori","year":"2011","journal-title":"Int J Poult Sci."},{"key":"key2.0261002114004e+13_B4","doi-asserted-by":"crossref","first-page":"813","DOI":"10.5187\/jast.2022.e56","article-title":"Current advances in detection of abnormal egg: a\n                        review","volume":"64","author":"So","year":"2022","journal-title":"J Anim Sci Technol."},{"key":"key2.0261002114004e+13_B5","doi-asserted-by":"crossref","first-page":"5546","DOI":"10.3390\/s20195546","article-title":"Non-destructive assessment of chicken egg\n                        fertility","volume":"20","author":"Adegbenjo","year":"2020","journal-title":"Sensors."},{"key":"key2.0261002114004e+13_B6","doi-asserted-by":"crossref","first-page":"289","DOI":"10.1079\/WPS200417","article-title":"Non-destructive measurements of the egg quality","volume":"60","author":"De Ketelaere","year":"2004","journal-title":"World\u2019s Poult Sci J."},{"key":"key2.0261002114004e+13_B7","doi-asserted-by":"crossref","first-page":"2354","DOI":"10.3390\/ani13142354","article-title":"A computer vision-based automatic system for egg grading and\n                        defect detection","volume":"13","author":"Yang","year":"2023","journal-title":"Animals."},{"key":"key2.0261002114004e+13_B8","doi-asserted-by":"crossref","first-page":"285","DOI":"10.33096\/ilkom.v13i3.937.285-293","article-title":"Identification of chicken egg fertility using SVM classifier\n                        based on first-order statistical feature extraction","volume":"13","author":"Saifullah","year":"2021","journal-title":"Ilk J Ilm."},{"key":"key2.0261002114004e+13_B9","doi-asserted-by":"crossref","unstructured":"Saifullah\n                            S\n                        \n                        \n                            Permadi\n                            VA\n                        \n                    \n                    Comparison of egg fertility identification based on GLCM feature\n                        extraction using backpropagation and K-means clustering algorithms\n                    In\n                    Proceedings of the 2019 5th International Conference on Science in\n                        Information Technology (ICSITech)\n                    2019\n                    10.1109\/ICSITech46713.2019.8987496","DOI":"10.1109\/ICSITech46713.2019.8987496"},{"key":"key2.0261002114004e+13_B10","doi-asserted-by":"crossref","first-page":"574","DOI":"10.3390\/vetsci9100574","article-title":"Deep learning based egg fertility detection","volume":"9","author":"\u00c7evik","year":"2022","journal-title":"Vet Sci."},{"key":"key2.0261002114004e+13_B11","article-title":"Squeezenet: AlexNet-level accuracy with 50x fewer parameters and\n                        <0.5MB model size","author":"Iandola","year":"2016","journal-title":"arXiv preprint arXiv:1602.07360"},{"key":"key2.0261002114004e+13_B12","volume-title":"ShuffleNet: an extremely efficient convolutional neural network for\n                        mobile devices","author":"Zhang","year":"2018"},{"key":"key2.0261002114004e+13_B13","doi-asserted-by":"crossref","unstructured":"Sandler\n                            M\n                        \n                        \n                            Howard\n                            A\n                        \n                        \n                            Zhu\n                            M\n                        \n                        \n                            Zhmoginov\n                            A\n                        \n                        \n                            Chen\n                            LC\n                        \n                    \n                    MobileNetV2: inverted residuals and linear bottlenecks\n                    In\n                    Proceedings of the 2018 IEEE\/CVF Conference on Computer Vision and\n                        Pattern Recognition\n                    2018\n                    10.1109\/CVPR.2018.00474","DOI":"10.1109\/CVPR.2018.00474"},{"key":"key2.0261002114004e+13_B14","doi-asserted-by":"crossref","unstructured":"Zoph\n                            B\n                        \n                        \n                            Vasudevan\n                            V\n                        \n                        \n                            Shlens\n                            J\n                        \n                        \n                            Le\n                            QV\n                        \n                    \n                    Learning transferable architectures for scalable image\n                        recognition\n                    In\n                    Proceedings of the 2018 IEEE\/CVF Conference on Computer Vision and\n                        Pattern Recognition\n                    2018\n                    10.1109\/CVPR.2018.00907","DOI":"10.1109\/CVPR.2018.00907"},{"key":"key2.0261002114004e+13_B15","unstructured":"Tan\n                            M\n                        \n                        \n                            Le\n                            QV\n                        \n                    \n                    EfficientNET: rethinking model scaling for convolutional neural\n                        networks\n                    In\n                    Proceedings of the 36th International Conference on Machine\n                        Learning\n                    2019"},{"key":"key2.0261002114004e+13_B16","doi-asserted-by":"crossref","unstructured":"Szegedy\n                            C\n                        \n                        \n                            Liu\n                            W\n                        \n                        \n                            Jia\n                            Y\n                        \n                        \n                            Sermanet\n                            P\n                        \n                        \n                            Reed\n                            S\n                        \n                        \n                            Anguelov\n                            D\n                        \n                        \n                    \n                    Going deeper with convolutions\n                    In\n                    Proceedings of the 2015 IEEE conference on computer vision and\n                        pattern recognition (CVPR)\n                    2015\n                    10.1109\/CVPR.2015.7298594","DOI":"10.1109\/CVPR.2015.7298594"},{"key":"key2.0261002114004e+13_B17","doi-asserted-by":"crossref","unstructured":"Szegedy\n                            C\n                        \n                        \n                            Vanhoucke\n                            V\n                        \n                        \n                            Ioffe\n                            S\n                        \n                        \n                            Shlens\n                            J\n                        \n                        \n                            Wojna\n                            Z\n                        \n                    \n                    Rethinking the inception architecture for computer vision\n                    In\n                    Proceedings of the 2016 IEEE Conference on Computer Vision and\n                        Pattern Recognition\n                    2016\n                    10.1109\/CVPR.2016.308","DOI":"10.1109\/CVPR.2016.308"},{"key":"key2.0261002114004e+13_B18","doi-asserted-by":"crossref","unstructured":"He\n                            K\n                        \n                        \n                            Zhang\n                            X\n                        \n                        \n                            Ren\n                            S\n                        \n                        \n                            Sun\n                            J\n                        \n                    \n                    Deep residual learning for image recognition\n                    In\n                    Proceedings of the 2016 IEEE Conference on Computer Vision and\n                        Pattern Recognition (CVPR)\n                    2016\n                    10.1109\/CVPR.2016.90\n                    26180094","DOI":"10.1109\/CVPR.2016.90"},{"key":"key2.0261002114004e+13_B19","doi-asserted-by":"crossref","first-page":"84","DOI":"10.1145\/3065386","article-title":"ImageNet classification with deep convolutional neural\n                        networks","volume":"60","author":"Krizhevsky","year":"2017","journal-title":"Commun ACM."},{"key":"key2.0261002114004e+13_B20","article-title":"Very deep convolutional networks for large-scale image\n                        recognition","author":"Simonyan","year":"2014","journal-title":"arXiv preprint arXiv:1409.1556 [Preprint]"},{"key":"key2.0261002114004e+13_B21","unstructured":"MathWorks\n                    \n                    Deep learning toolbox documentation [Internet]\n                    MathWorks Inc.\n                    2026\n                    [cited 2025 Jun 12]\n                    https:\/\/mathworks.com\/help\/deeplearning\/index.html"},{"key":"key2.0261002114004e+13_B22","doi-asserted-by":"crossref","first-page":"854","DOI":"10.26555\/jiteki.v9i3.26722","article-title":"Nondestructive chicken egg fertility detection using CNN-transfer\n                        learning algorithms","volume":"9","author":"Saifullah","year":"2023","journal-title":"J Ilm Tek Elektro Komput Inform."},{"key":"key2.0261002114004e+13_B23","doi-asserted-by":"crossref","first-page":"2900","DOI":"10.3934\/era.2023146","article-title":"Brain tumor detection using CNN, AlexNet & GoogLeNet\n                        ensembling learning approaches","volume":"31","author":"Swarup","year":"2023","journal-title":"Electron Res Arch."},{"key":"key2.0261002114004e+13_B24","doi-asserted-by":"crossref","first-page":"997","DOI":"10.3390\/electronics8090997","article-title":"Evolutionary-fuzzy-integral-based convolutional neural networks\n                        for facial image classification","volume":"8","author":"Lin","year":"2019","journal-title":"Electronics."}],"container-title":["Journal of Animal Science and Technology"],"original-title":[],"language":"en","link":[{"URL":"http:\/\/www.ejast.org\/download\/download_pdf?doi=10.5187\/jast.2600100","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"http:\/\/www.ejast.org\/download\/download_pdf?doi=10.5187\/jast.2600100","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,2]],"date-time":"2026-10-02T03:00:14Z","timestamp":1790910014000},"score":1,"resource":{"primary":{"URL":"http:\/\/www.ejast.org\/archive\/view_article?doi=10.5187\/jast.2600100"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7]]},"references-count":24,"journal-issue":{"issue":"4","published-online":{"date-parts":[[2026,7,31]]},"published-print":{"date-parts":[[2026,7]]}},"alternative-id":["10.5187\/jast.2600100"],"URL":"https:\/\/doi.org\/10.5187\/jast.2600100","relation":{},"ISSN":["2672-0191","2055-0391"],"issn-type":[{"value":"2672-0191","type":"print"},{"value":"2055-0391","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7]]},"assertion":[{"value":"2026-03-16","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-05-06","order":1,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}}]}}