<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Publishing DTD v1.3 20210610//EN" "JATS-journalpublishing1-3.dtd">
<article article-type="research-article" dtd-version="1.3" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xml:lang="ru"><front><journal-meta><journal-id journal-id-type="publisher-id">vsgiu</journal-id><journal-title-group><journal-title xml:lang="ru">Вестник Сибирского государственного индустриального университета</journal-title><trans-title-group xml:lang="en"><trans-title>Bulletin of the Siberian State Industrial University</trans-title></trans-title-group></journal-title-group><issn pub-type="ppub">2304 - 4497</issn><issn pub-type="epub">2307-1710</issn><publisher><publisher-name>Федеральное государственное бюджетное образовательное учреждение высшего образования "Сибирский государственный индустриальный университет"</publisher-name></publisher></journal-meta><article-meta><article-id pub-id-type="doi">10.57070/2304-4497-2025-2(52)-85-91</article-id><article-id custom-type="elpub" pub-id-type="custom">vsgiu-259</article-id><article-categories><subj-group subj-group-type="heading"><subject>Research Article</subject></subj-group><subj-group subj-group-type="section-heading" xml:lang="ru"><subject>Раздел 2. Металлургия и материаловедение</subject></subj-group><subj-group subj-group-type="section-heading" xml:lang="en"><subject>Section 2. Metallurgy and Materials Science</subject></subj-group></article-categories><title-group><article-title>РАСПОЗНАВАНИЕ ДЕФЕКТОВ НА ПОВЕРХНОСТИ МЕТАЛЛА С ИСПОЛЬЗОВАНИЕМ МЕТОДА МАШИННОГО ОБУЧЕНИЯ</article-title><trans-title-group xml:lang="en"><trans-title>RECOGNITION OF DEFECTS ON THE METAL SURFACE USING MACHINE LEARNING</trans-title></trans-title-group></title-group><contrib-group><contrib contrib-type="author" corresp="yes"><contrib-id contrib-id-type="orcid">https://orcid.org/0009-0007-5845-4928</contrib-id><name-alternatives><name name-style="eastern" xml:lang="ru"><surname>Кузнецова</surname><given-names>Валентина Анатольевна</given-names></name><name name-style="western" xml:lang="en"><surname>Kuznetsova</surname><given-names>Valentina A.</given-names></name></name-alternatives><bio xml:lang="ru"><p>обучающийся кафедры прикладной математики и информатики</p></bio><bio xml:lang="en"><p>student of the department of  applied mathematics and computer science</p></bio><email xlink:type="simple">valyakuz28@mail.ru</email><xref ref-type="aff" rid="aff-1"/></contrib><contrib contrib-type="author" corresp="yes"><contrib-id contrib-id-type="orcid">https://orcid.org/0000-0002-4566-528X</contrib-id><name-alternatives><name name-style="eastern" xml:lang="ru"><surname>Маркидонов</surname><given-names>Артем Владимирович</given-names></name><name name-style="western" xml:lang="en"><surname>Markidonov</surname><given-names>Artem V.</given-names></name></name-alternatives><bio xml:lang="ru"><p>д.ф.-м.н., доцент, заведующий кафедрой информатики и вычислительной техники им. В.К. Буторина, профессор кафедры  прикладной математики и информатики</p></bio><bio xml:lang="en"><p>Dr. Sci. (Phys.-math.), Associate Professor, head of the department of ivt after. professor of the department of applied mathematics and computer science</p></bio><email xlink:type="simple">markidonov_artem@mail.ru</email><xref ref-type="aff" rid="aff-2"/></contrib></contrib-group><aff-alternatives id="aff-1"><aff xml:lang="ru"><institution>Сибирский государственный индустриальный университет</institution><country>Россия</country></aff><aff xml:lang="en"><institution>Siberian  State Industrial University</institution><country>Russian Federation</country></aff></aff-alternatives><aff-alternatives id="aff-2"><aff xml:lang="ru"><institution>Кузбасский гуманитарно-педагогический институт Кемеровского  государственного университета; Сибирский  государственный индустриальный университет</institution><country>Russian Federation</country></aff><aff xml:lang="en"><institution>V.K. Butorin, Kuzbass Humanitarian and Pedagogical Institute of Kemerovo State University, Siberian State Industrial University</institution><country>Russian Federation</country></aff></aff-alternatives><pub-date pub-type="collection"><year>2025</year></pub-date><pub-date pub-type="epub"><day>07</day><month>07</month><year>2025</year></pub-date><volume>0</volume><issue>2</issue><fpage>85</fpage><lpage>91</lpage><permissions><copyright-statement>Copyright &amp;#x00A9; Кузнецова В.А., Маркидонов А.В., 2025</copyright-statement><copyright-year>2025</copyright-year><copyright-holder xml:lang="ru">Кузнецова В.А., Маркидонов А.В.</copyright-holder><copyright-holder xml:lang="en">Kuznetsova V., Markidonov A.</copyright-holder><license xml:lang="ru" license-type="creative-commons-attribution" xlink:href="https://creativecommons.org/licenses/by/4.0/" xlink:type="simple"><license-p>Данная работа распространяется под лицензией Creative Commons Attribution 4.0.</license-p></license><license xml:lang="en" license-type="creative-commons-attribution" xlink:href="https://creativecommons.org/licenses/by/4.0/" xlink:type="simple"><license-p>This work is licensed under a Creative Commons Attribution 4.0 License.</license-p></license></permissions><self-uri xlink:href="https://vestnik.sibsiu.ru/jour/article/view/259">https://vestnik.sibsiu.ru/jour/article/view/259</self-uri><abstract><p>В связи с увеличением требований к качеству продукции в металлургической и машиностроительной промышленностях возникает необходимость внедрения современных технологий для автоматического контроля качества. Поверхностные дефекты металлических изделий (трещины, царапины и включения) напрямую влияют на надежность и долговечность продукции. Традиционные методы визуального и оптического контроля требуют значительных временных и трудовых затрат, подвержены влиянию человеческого фактора и не всегда обеспечивают достаточную точность. В рамках исследования был проведен обзор современных публикаций, в которых рассматриваются подходы к автоматической классификации дефектов, а также обсуждаются возможности и ограничения архитектур нейронных сетей. Анализ источников позволил выявить тенденции развития в рассматриваемой области и обосновать выбор архитектуры модели. Предлагается подход к распознаванию дефектов на изображениях металлических поверхностей с использованием сверточных нейронных сетей. Разработана архитектура модели, включающая три сверточных слоя и полносвязные нейроны, оптимизированные с использованием функции активации ReLU, слоя Dropout и выходного слоя Softmax. Для обучения модели использовали открытый набор данных, содержащий 1800 черно-белых изображений с шестью различными типами дефектов. Точность классификации составила 95,83 %, а значение функции потерь ‒ 0,0862. При проверке на тестовой выборке модель верно распознала 70 из 72 изображений. Проведенное исследование подтверждает эффективность нейронных сетей в задаче обнаружения визуальных дефектов. Представленная модель может быть использована в системах автоматизированного контроля качества и дополнительно адаптирована под различные промышленные условия. В дальнейшем планируется оптимизация архитектуры модели для повышения устойчивости к шумам и вариативности данных.</p></abstract><trans-abstract xml:lang="en"><p>Due to the increase in product quality requirements in the metallurgical and machine building industries, it is necessary to introduce modern technologies for automatic quality control. Surface defects of metal products (cracks, scratches and inclusions) directly affect the reliability and durability of products. Traditional methods of visual and optical control require significant time and labor costs, are subject to the influence of the human factor and do not always provide sufficient accuracy. Within the framework of the study, a review of modern publications was conducted, which consider approaches to automatic defect classification, as well as discuss the possibilities and limitations of neural network architectures. The analysis of the sources made it possible to identify development trends in the field under consideration and justify the choice of the model architecture. An approach to the detection of defects in images of metal surfaces using convolutional neural networks is proposed. The architecture of the model has been developed, which includes three convolutional layers and fully connected neurons optimized using the ReLU activation function, the Dropout layer and the Softmax output layer. To train the model, we used an open dataset containing 1800 black and white images with six different types of defects. The classification accuracy was 95.83 %, and the value of the loss function was 0.0862. When tested on a test sample, the model correctly recognized 70 out of 72 images. The conducted research confirms the effectiveness of neural networks in the task of detecting visual defects. The presented model can be used in automated quality control systems and additionally adapted to various industrial conditions. In the future, optimization of the model architecture is planned to increase noise tolerance and data variability.</p></trans-abstract><kwd-group xml:lang="ru"><kwd>машинное обучение</kwd><kwd>дефекты металлических поверхностей</kwd><kwd>сверточная нейронная сеть</kwd><kwd>классификация изображений</kwd></kwd-group><kwd-group xml:lang="en"><kwd>machine learning</kwd><kwd>defects of metal surfaces</kwd><kwd>convolutional neural network</kwd><kwd>image classification</kwd></kwd-group></article-meta></front><back><ref-list><title>References</title><ref id="cit1"><label>1</label><citation-alternatives><mixed-citation xml:lang="ru">Huang Y., Yu T., Wan K., Yuan J. Detection and classification of metal workpiece surface defects based on machine vision. 2021 IEEE International Conference on Advances in Electrical Engineering and Computer Applications (AEECA). 2021:983–987.</mixed-citation><mixed-citation xml:lang="en">Huang Y., Yu T., Wan K., Yuan J. Detection and classification of metal workpiece surface defects based on machine vision. 2021 IEEE International Conference on Advances in Electrical Engineering and Computer Applications (AEECA). 2021: 983–987.</mixed-citation></citation-alternatives></ref><ref id="cit2"><label>2</label><citation-alternatives><mixed-citation xml:lang="ru">https://doi.org/10.1109/AEECA52519.2021.9574344</mixed-citation><mixed-citation xml:lang="en">https://doi.org/10.1109/AEECA52519.2021.9574344</mixed-citation></citation-alternatives></ref><ref id="cit3"><label>3</label><citation-alternatives><mixed-citation xml:lang="ru">Bai J., Wu D., Shelley T., Schubel P., Twine D., Russell J., Zeng X., Zhang J. A comprehensive survey on machine learning driven material defect detection: challenges, solutions, and future prospects. ACM Computing Surveys. 2024. https://doi.org/10.48550/arXiv.2406.07880</mixed-citation><mixed-citation xml:lang="en">Bai J., Wu D., Shelley T., Schubel P., Twine D., Russell J., Zeng X., Zhang J. A comprehensive survey on machine learning driven material defect detection: challenges, solutions, and future prospects. ACM Computing Surveys. 2024. https://doi.org/10.48550/arXiv.2406.07880</mixed-citation></citation-alternatives></ref><ref id="cit4"><label>4</label><citation-alternatives><mixed-citation xml:lang="ru">Zhou C., Lu Z., Lv Z. et al. Metal surface defect detection based on improved YOLOv5. Scientific Reports. 2023;13:20803.</mixed-citation><mixed-citation xml:lang="en">Zhou C., Lu Z., Lv Z. et al. Metal surface defect detection based on improved YOLOv5. Scientific Reports. 2023;13:20803. https://doi.org/10.1038/s41598-023-47716-2</mixed-citation></citation-alternatives></ref><ref id="cit5"><label>5</label><citation-alternatives><mixed-citation xml:lang="ru">https://doi.org/10.1038/s41598-023-47716-2</mixed-citation><mixed-citation xml:lang="en">Chen S., Zhou F., Gao G., Ge X., Wang R. Unleashing the power of AI in detecting metal surface defects: an optimized YOLOv7-tiny model approach. PeerJ. Computer science. 2024;10:e1727. https://doi.org/10.7717/peerj-cs.1727</mixed-citation></citation-alternatives></ref><ref id="cit6"><label>6</label><citation-alternatives><mixed-citation xml:lang="ru">Chen S., Zhou F., Gao G., Ge X., Wang R. Unleashing the power of AI in detecting metal surface defects: an optimized YOLOv7-tiny model approach. PeerJ. Computer science. 2024;10:e1727. https://doi.org/10.7717/peerj-cs.1727</mixed-citation><mixed-citation xml:lang="en">Huang Y.C., Hung K.C., Lin J.C. Automated machine learning system for defect detection on cylindrical metal surfaces. Sensors. 2022;22(24):9783. https://doi.org/10.3390/s22249783</mixed-citation></citation-alternatives></ref><ref id="cit7"><label>7</label><citation-alternatives><mixed-citation xml:lang="ru">Huang Y.C., Hung K.C., Lin J.C. Automated machine learning system for defect detection on cylindrical metal surfaces. Sensors. 2022;22(24):9783. https://doi.org/10.3390/s22249783</mixed-citation><mixed-citation xml:lang="en">Baleev I.A., Zemtsov A.N., Zybin M.I., Smirnov V.A. Recognizing defects in metal alloys using OpenCV computer vision algo-rithms. Inzhenernyi vestnik Dona. 2021; 3 (75):78–87. (In Russ.).</mixed-citation></citation-alternatives></ref><ref id="cit8"><label>8</label><citation-alternatives><mixed-citation xml:lang="ru">Балеев И.А., Земцов А.Н., Зыбин М.И., Смирнов В.А. Распознавание дефектов на металлических сплавах с помощью алгоритмов компьютерного зрения OpenCV. Инженерный вестник Дона. 2021;3(75):78–87.</mixed-citation><mixed-citation xml:lang="en">Pogadaeva E.Yu. Recognition of welded joint de-fects using a photographic image for visual inspec-tion. Molodoi ucheny. 2020; 43(333):5–9. (In Russ.).</mixed-citation></citation-alternatives></ref><ref id="cit9"><label>9</label><citation-alternatives><mixed-citation xml:lang="ru">Погадаева Е.Ю. Распознавание дефектов сварных соединений по фотоизображению для проведения визуального контроля. Молодой ученый. 2020;43(333):5–9.</mixed-citation><mixed-citation xml:lang="en">Rumanovskii I.G., Kalinnikov N. A., Nikitin N.A. Application of neural network technologies for flaw detection of railway tracks. Vestnik Tikhookeanskogo gosudarstvennogo universiteta. 2023; 4(71):25–40. (In Russ.).</mixed-citation></citation-alternatives></ref><ref id="cit10"><label>10</label><citation-alternatives><mixed-citation xml:lang="ru">Румановский И.Г., Калинников Н. А., Никитин Н.А. Применение нейросетевых технологий для дефектоскопии железнодорожных путей. Вестник Тихоокеанского государственного университета. 2023; 4(71):25–40.</mixed-citation><mixed-citation xml:lang="en">Dean K. Argonne scientists use AI to detect hidden defects in stainless steel. Nuclear News, 2025. URL: https://www.ans.org/news/article-6706/argonne-scientists-use-ai-to-detect-hidden-defects-in-stainless-steel. (Дата обращения: 10.04.2025).</mixed-citation></citation-alternatives></ref><ref id="cit11"><label>11</label><citation-alternatives><mixed-citation xml:lang="ru">Dean K. Argonne scientists use AI to detect hidden defects in stainless steel. Nuclear News, 2025. URL: https://www.ans.org/news/article-6706/argonne-scientists-use-ai-to-detect-hidden-defects-in-stainless-steel. (Дата обращения: 10.04.2025).</mixed-citation><mixed-citation xml:lang="en">Panova V.S., Kuznetsova V.A., Panchenko I.A. Application of neural networks for predicting the properties of high-entropy alloys. In: Design Ultrafine-grained and nanostructured materials: Collection of works of the open school-conference of the CIS countries. Ufa University of Science and Technology. 2024:154. (In Russ.).</mixed-citation></citation-alternatives></ref><ref id="cit12"><label>12</label><citation-alternatives><mixed-citation xml:lang="ru">Панова В.С., Кузнецова В.А., Панченко И.А. Применение нейронных сетей для прогнозирования свойств высокоэнтропийных сплавов. В кн.: Ультрамелкозернистые и наноструктурные материалы: Сборник трудов открытой школы-конференции стран СНГ. Уфимский университет науки и технологий. 2024:154.</mixed-citation><mixed-citation xml:lang="en">Artificial Intelligence in Metallurgy: How It Is Used to Detect Defects. 2024. (In Russ.). URL: https://indpages.ru/prom/iskusstvennyj-intellekt-v-metallurgii-kak-ego-ispolzuyut-dlya-obnaruzheniya-defektov.</mixed-citation></citation-alternatives></ref><ref id="cit13"><label>13</label><citation-alternatives><mixed-citation xml:lang="ru">Искусственный интеллект в металлургии: как его используют для обнаружения де-фектов. 2024. URL: https://indpages.ru/prom/ iskusstvennyj-intellekt-v-metallurgii-kak-ego-ispolzuyut-dlya-obnaruzheniya-defektov/ (Дата обращения: 10.04.2025).</mixed-citation><mixed-citation xml:lang="en">Wang S., Xia X., Ye L., Yang B. Automatic detection and classification of steel surface defect using deep convolutional neural networks. Metals. 2021;11(3):388.</mixed-citation></citation-alternatives></ref><ref id="cit14"><label>14</label><citation-alternatives><mixed-citation xml:lang="ru">Wang S., Xia X., Ye L., Yang B. Automatic detection and classification of steel surface defect using deep convolutional neural networks. Metals. 2021;11(3):388.</mixed-citation><mixed-citation xml:lang="en">https://doi.org/10.3390/met11030388</mixed-citation></citation-alternatives></ref><ref id="cit15"><label>15</label><citation-alternatives><mixed-citation xml:lang="ru">https://doi.org/10.3390/met11030388</mixed-citation><mixed-citation xml:lang="en">Suh S. Optimal surface defect detector design based on deep learning for 3D geometry. Scientific Reports. 2025;15:5527.</mixed-citation></citation-alternatives></ref><ref id="cit16"><label>16</label><citation-alternatives><mixed-citation xml:lang="ru">Suh S. Optimal surface defect detector design based on deep learning for 3D geometry. Scientific Reports. 2025;15:5527.</mixed-citation><mixed-citation xml:lang="en">https://doi.org/10.1038/s41598-025-88112-2</mixed-citation></citation-alternatives></ref><ref id="cit17"><label>17</label><citation-alternatives><mixed-citation xml:lang="ru">https://doi.org/10.1038/s41598-025-88112-2</mixed-citation><mixed-citation xml:lang="en">Keshinro B. Image detection and classification: a machine learning approach. 2022. http://dx.doi.org/10.2139/ssrn.4281011</mixed-citation></citation-alternatives></ref><ref id="cit18"><label>18</label><citation-alternatives><mixed-citation xml:lang="ru">Keshinro B. Image detection and classification: a machine learning approach. 2022. http://dx.doi.org/10.2139/ssrn.4281011</mixed-citation><mixed-citation xml:lang="en">Rybakov K.M., Khamitov R.M. Problems of surface flaw detection of metals using machine learning and ways to solve them. International Journal of Advanced Studies. 2024; 14(1):196–204.       (In Russ.).</mixed-citation></citation-alternatives></ref><ref id="cit19"><label>19</label><citation-alternatives><mixed-citation xml:lang="ru">Рыбаков К.М., Хамитов Р.М. Проблемы поверхностной дефектоскопии металлов с использованием машинного обучения и пути их решения. International Journal of Advanced Studies. 2024; 14(1):196–204. https://doi.org./10.12731/2227-930X-2024-14-1-289</mixed-citation><mixed-citation xml:lang="en">https://doi.org./10.12731/2227-930X-2024-14-1-289</mixed-citation></citation-alternatives></ref><ref id="cit20"><label>20</label><citation-alternatives><mixed-citation xml:lang="ru">Metal Surface Defects Dataset. Kaggle. URL: https://www.kaggle.com/datasets/fantacher/neu-metal-surface-defects-data/data?select =NEU+Metal+Surface+Defects+Data (Дата обращения: 25.03.2025).</mixed-citation><mixed-citation xml:lang="en">Metal Surface Defects Dataset. Kaggle. URL: https://www.kaggle.com/datasets/fantacher/neu-metal-surface-defects-data/data?select =NEU+Metal+Surface+Defects+Data (Дата обращения: 25.03.2025).</mixed-citation></citation-alternatives></ref><ref id="cit21"><label>21</label><citation-alternatives><mixed-citation xml:lang="ru">Lv Q., Zhang S., Wang Y. Deep Learning Model of Image Classification Using Machine Learning. Advances in Multimedia. 2022; 3351256. https://doi.org/10.1155/2022/3351256</mixed-citation><mixed-citation xml:lang="en">Lv Q., Zhang S., Wang Y. Deep Learning Model of Image Classification Using Machine Learning. Advances in Multimedia. 2022; 3351256.  https://doi.org/10.1155/2022/3351256</mixed-citation></citation-alternatives></ref><ref id="cit22"><label>22</label><citation-alternatives><mixed-citation xml:lang="ru">Pilyay A. I. Detection of defects in building materials using artificial intelligence systems. Construction and Architecture. 2023;11(1):20. https://doi.org/10.29039/2308-0191-2022-11-1-20-20</mixed-citation><mixed-citation xml:lang="en">Pilyay A. I. Detection of defects in building materials using artificial intelligence systems. Construction and Architecture. 2023;11(1):20. https://doi.org/10.29039/2308-0191-2022-11-1-20-20</mixed-citation></citation-alternatives></ref><ref id="cit23"><label>23</label><citation-alternatives><mixed-citation xml:lang="ru">Cherkasov N., Ivanov S., Ivanov M., Ulanov A. Detection of defects in welded butt joints based on laser scanning: neural networks approach. In: International Ural conference on electrical power engineering (UralCon). 2023: 775–779. https://doi.org/10.1109/UralCon59258.2023.10291060</mixed-citation><mixed-citation xml:lang="en">Cherkasov N., Ivanov S., Ivanov M., Ulanov A. Detection of defects in welded butt joints based on laser scanning: neural networks approach. In: International Ural conference on electrical power engineering (UralCon). 2023:775–779. https://doi.org/10.1109/UralCon59258.2023.10291060</mixed-citation></citation-alternatives></ref><ref id="cit24"><label>24</label><citation-alternatives><mixed-citation xml:lang="ru">Al-Mamun A.M., Hossain M.R., Sharmin M.M. Detection and classification of metal surface defects using lite convolutional neural network (LCNN). Material Science &amp; Engineering International Journal. 2024; 8(3): 72–76.</mixed-citation><mixed-citation xml:lang="en">Al-Mamun A.M., Hossain M.R., Sharmin M.M. Detection and classification of metal surface defects using lite convolutional neural network (LCNN). Material Science &amp; Engineering International Journal. 2024; 8(3): 72–76.</mixed-citation></citation-alternatives></ref><ref id="cit25"><label>25</label><citation-alternatives><mixed-citation xml:lang="ru">https://doi.org/10.15406/mseij.2024.08.00239</mixed-citation><mixed-citation xml:lang="en">https://doi.org/10.15406/mseij.2024.08.00239</mixed-citation></citation-alternatives></ref></ref-list><fn-group><fn fn-type="conflict"><p>The authors declare that there are no conflicts of interest present.</p></fn></fn-group></back></article>
