<?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">ntv</journal-id><journal-title-group><journal-title xml:lang="ru">Научно-технический вестник информационных технологий, механики и оптики</journal-title><trans-title-group xml:lang="en"><trans-title>Scientific and Technical Journal of Information Technologies, Mechanics and Optics</trans-title></trans-title-group></journal-title-group><issn pub-type="ppub">2226-1494</issn><issn pub-type="epub">2500-0373</issn><publisher><publisher-name>Университет ИТМО</publisher-name></publisher></journal-meta><article-meta><article-id pub-id-type="doi">10.17586/2226-1494-2026-26-2-420-427</article-id><article-id custom-type="elpub" pub-id-type="custom">ntv-605</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>МАТЕМАТИЧЕСКОЕ И КОМПЬЮТЕРНОЕ МОДЕЛИРОВАНИЕ</subject></subj-group><subj-group subj-group-type="section-heading" xml:lang="en"><subject>MODELING AND SIMULATION</subject></subj-group></article-categories><title-group><article-title>Применение гибридных методов искусственного интеллекта для практических производственных задач в условиях труднодоступности обучающих данных</article-title><trans-title-group xml:lang="en"><trans-title>Application of hybrid artificial intelligence methods to practical industrial tasks under conditions of scarce training data</trans-title></trans-title-group></title-group><contrib-group><contrib contrib-type="author" corresp="yes"><contrib-id contrib-id-type="orcid">https://orcid.org/0009-0005-5479-2482</contrib-id><name-alternatives><name name-style="eastern" xml:lang="ru"><surname>Соловьев</surname><given-names>Г. В.</given-names></name><name name-style="western" xml:lang="en"><surname>Solovev</surname><given-names>G. V.</given-names></name></name-alternatives><bio xml:lang="ru"><p>Соловьев Глеб Витальевич — аспирант, инженер-исследователь</p><p>Санкт-Петербург, 197101</p><p>sc 59409934700</p></bio><bio xml:lang="en"><p>Gleb V. Solovev — PhD Student, Engineer Researcher</p><p>Saint Petersburg, 197101</p><p>sc 59409934700</p></bio><email xlink:type="simple">glebsolo46@gmail.com</email><xref ref-type="aff" rid="aff-1"/></contrib><contrib contrib-type="author" corresp="yes"><contrib-id contrib-id-type="orcid">https://orcid.org/0000-0003-4413-3676</contrib-id><name-alternatives><name name-style="eastern" xml:lang="ru"><surname>Соколов</surname><given-names>М. С.</given-names></name><name name-style="western" xml:lang="en"><surname>Sokolov</surname><given-names>M. S.</given-names></name></name-alternatives><bio xml:lang="ru"><p>Соколов Михаил Сергеевич — кандидат технических наук, ведущий научный сотрудник</p><p>Санкт-Петербург, 197101</p><p>sc 56994304500</p></bio><bio xml:lang="en"><p>Mikhail S. Sokolov — PhD, Leading Researcher</p><p>Saint Petersburg, 197101</p><p>sc 56994304500</p></bio><email xlink:type="simple">mikhail.sokolov@itmo.ru</email><xref ref-type="aff" rid="aff-1"/></contrib></contrib-group><aff-alternatives id="aff-1"><aff xml:lang="ru"><institution>Университет ИТМО</institution><country>Россия</country></aff><aff xml:lang="en"><institution>ITMO University</institution><country>Russian Federation</country></aff></aff-alternatives><pub-date pub-type="collection"><year>2026</year></pub-date><pub-date pub-type="epub"><day>20</day><month>04</month><year>2026</year></pub-date><volume>26</volume><issue>2</issue><fpage>420</fpage><lpage>427</lpage><permissions><copyright-statement>Copyright &amp;#x00A9; Соловьев Г.В., Соколов М.С., 2026</copyright-statement><copyright-year>2026</copyright-year><copyright-holder xml:lang="ru">Соловьев Г.В., Соколов М.С.</copyright-holder><copyright-holder xml:lang="en">Solovev G.V., Sokolov M.S.</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://ntv.elpub.ru/jour/article/view/605">https://ntv.elpub.ru/jour/article/view/605</self-uri><abstract><p>Введение. Современные производственные задачи, такие как контроль качества лазерной сварки и локализация геометрических признаков в промышленных процессах, требуют применения инновационных подходов машинного обучения. Недостаток размеченных данных и сложность геометрической аннотации являются критическими барьерами при разработке автоматизированных систем контроля. Научная новизна предлагаемого подхода заключается в комплексном использовании гибридных методов, объединяющих эволюционную оптимизацию, диффузионные модели и сверточные нейронные сети для эффективного решения практических инженерных задач с ограниченными ресурсами данных. Метод. Предложенный подход состоит из двух интегрированных компонентов. Первый компонент реализует гибридный алгоритм генерации синтетических данных, объединяющий эволюционную оптимизацию для генерации разнообразных геометрических вариантов с использованием диффузионных моделей для синтеза фотореалистичных изображений. Второй компонент включает специализированную архитектуру глубокого обучения, оптимизированную для точной локализации и классификации геометрических признаков в производственных контекстах. Обучение реализуется с применением комбинированной функции потерь, интегрирующей регрессионные и классификационные критерии. Основные результаты. На примере задачи контроля качества лазерной сварки синтетический набор данных расширен с 120 оригинальных изображений до 4537 реалистичных примеров, что позволило повысить точность сегментации швов с 2,4 до 0,75 по метрике потерь box loss. Для задачи локализации координат шва достигнута ошибка предсказания 31,8 пикселов по оси Y и 3,3 пиксела по оси X на исходном разрешении 1024 × 2448 пикселов. Экспериментальное сравнение показало превосходство сверточных архитектур над трансформерными моделями при сопоставимом количестве параметров, а также лучшую точность при регрессии с одного кадра, чем при использовании последовательности кадров. Обсуждение. Предложенные методы демонстрируют значительное превосходство по сравнению с классическими подходами масштабирования данных (mixup, cutmix) и чистыми подходами к диффузионному синтезу, которые требуют интенсивной подготовки наборов данных. Интеграция эволюционной оптимизации обеспечивает контролируемое разнообразие геометрических вариантов, а диффузионные модели гарантируют фотореалистичность синтезированных образцов. Данный гибридный подход имеет широкие перспективы применения в других промышленных секторах с ограниченной доступностью размеченных данных благодаря возможности построения полного конвейера синтеза труднодоступных промышленных данных, а затем их использования для обучения прикладных методов искусственного интеллекта, решающих целевые промышленные задачи.</p></abstract><trans-abstract xml:lang="en"><p>Modern industrial tasks, such as quality control in laser welding and the localization of geometric features in industrial processes, require the application of innovative machine learning approaches. The scarcity of annotated data and the complexity of geometric annotation are critical barriers to the development of automated inspection systems. The scientific novelty of the proposed approach lies in the comprehensive use of hybrid methods, combining evolutionary optimization, diffusion models, and convolutional neural networks to effectively address practical engineering tasks with limited data resources. The proposed framework consists of two integrated components. The first component implements a hybrid algorithm for synthetic data generation, merging evolutionary optimization for generating diverse geometric variants with diffusion models for synthesizing photorealistic images. The second component involves a specialized deep learning architecture optimized for the precise localization and classification of geometric features in industrial settings. Training is performed using a combined loss function that integrates regression and classification criteria. In the case of laser welding quality control, the synthetic dataset was expanded from 120 original images to 4,537 realistic samples. This augmentation improved weld seam segmentation accuracy, reducing the box loss metric from 2.4 to 0.75. For the task of localizing weld seam coordinates, a prediction error of 31.8 pixels along the Y-axis and 3.3 pixels along the X-axis was achieved at the original resolution of 1,024 × 2,448 pixels. Experimental comparisons showed that convolutional architectures outperformed transformer-based models with a comparable number of parameters, and that regression from a single frame yielded higher accuracy than using a sequence of frames. The proposed methods demonstrate significant superiority over classical data augmentation techniques (e.g., mixup, cutmix) and pure diffusionbased synthesis approaches which require intensive dataset preparation. The integration of evolutionary optimization ensures controlled diversity in geometric variants, while diffusion models guarantee the photorealism of synthesized samples. This hybrid approach holds broad potential for application in other industrial sectors with limited availability of annotated data, owing to its capability to construct a complete pipeline for synthesizing hard-to-obtain industrial data and subsequently using it to train applied Artificial Intelligence methods for solving targeted industrial problems.</p></trans-abstract><kwd-group xml:lang="ru"><kwd>гибридные методы машинного обучения</kwd><kwd>эволюционная оптимизация</kwd><kwd>диффузионные модели</kwd><kwd>синтетические данные</kwd><kwd>сверточные нейронные сети</kwd><kwd>локализация геометрических признаков</kwd><kwd>контроль качества производства</kwd><kwd>лазерная сварка</kwd><kwd>искусственный интеллект</kwd></kwd-group><kwd-group xml:lang="en"><kwd>hybrid machine learning methods</kwd><kwd>evolutionary optimization</kwd><kwd>diffusion models</kwd><kwd>synthetic data</kwd><kwd>convolutional neural networks</kwd><kwd>geometric feature localization</kwd><kwd>industrial quality control</kwd><kwd>laser welding</kwd><kwd>artificial intelligence</kwd></kwd-group><funding-group><funding-statement xml:lang="ru">Исследование выполнено за счет гранта Российского научного фонда № 25-11-00367, https://rscf.ru/project/25-11-00367/.</funding-statement><funding-statement xml:lang="en">The study was supported by the grant No. 25-11-00367 from the Russian Science Foundation, https://rscf.ru/project/25-11-00367/.</funding-statement></funding-group></article-meta></front><back><ref-list><title>References</title><ref id="cit1"><label>1</label><citation-alternatives><mixed-citation xml:lang="ru">Xiao S., Liu Z., Yan Z., Wang M. Grad-MobileNet: A gradient-based unsupervised learning method for laser welding surface defect classification // Sensors. 2023. V. 23. N 9. P. 4563. https://doi.org/10.3390/s23094563</mixed-citation><mixed-citation xml:lang="en">Xiao S., Liu Z., Yan Z., Wang M. Grad-MobileNet: A gradient-based unsupervised learning method for laser welding surface defect classification // Sensors. 2023. V. 23. N 9. P. 4563. https://doi.org/10.3390/s23094563</mixed-citation></citation-alternatives></ref><ref id="cit2"><label>2</label><citation-alternatives><mixed-citation xml:lang="ru">Vu T., Sun B., Yuan B., Ngai A., Li Y., Frahm J.-M. Supervision interpolation via lossmix: generalizing mixup for object detection and beyond // Proceedings of the AAAI Conference on Artificial Intelligence. 2024. V. 38. N 6. P. 5280–5288. https://doi.org/10.1609/aaai.v38i6.28335</mixed-citation><mixed-citation xml:lang="en">Vu T., Sun B., Yuan B., Ngai A., Li Y., Frahm J.-M. Supervision interpolation via lossmix: generalizing mixup for object detection and beyond // Proceedings of the AAAI Conference on Artificial Intelligence. 2024. V. 38. N 6. P. 5280–5288. https://doi.org/10.1609/aaai.v38i6.28335</mixed-citation></citation-alternatives></ref><ref id="cit3"><label>3</label><citation-alternatives><mixed-citation xml:lang="ru">Jin X., Zhu H., Li S., Wang Z., Liu Z., Tian J., et al. A survey on mixup augmentations and beyond // arXiv. 2024. arXiv:2409.05202. https://doi.org/10.48550/arXiv.2409.05202</mixed-citation><mixed-citation xml:lang="en">Jin X., Zhu H., Li S., Wang Z., Liu Z., Tian J., et al. A survey on mixup augmentations and beyond // arXiv. 2024. arXiv:2409.05202. https://doi.org/10.48550/arXiv.2409.05202</mixed-citation></citation-alternatives></ref><ref id="cit4"><label>4</label><citation-alternatives><mixed-citation xml:lang="ru">Capogrosso L., Girella F., Taioli F., Chiara M., Aqeel M., Fummi F., et al. Diffusion-based image generation for in-distribution data augmentation in surface defect detection // Proc. of the 19th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications. 2024. V. 2. P. 409–416. https://doi.org/10.5220/0012350400003660</mixed-citation><mixed-citation xml:lang="en">Capogrosso L., Girella F., Taioli F., Chiara M., Aqeel M., Fummi F., et al. Diffusion-based image generation for in-distribution data augmentation in surface defect detection // Proc. of the 19th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications. 2024. V. 2. P. 409–416. https://doi.org/10.5220/0012350400003660</mixed-citation></citation-alternatives></ref><ref id="cit5"><label>5</label><citation-alternatives><mixed-citation xml:lang="ru">Tai Y., Yang K., Peng T., Huang Z., Zhang Z. Defect image sample generation with diffusion prior for steel surface defect recognition // IEEE Transactions on Automation Science and Engineering. 2024. V. 22. P. 8239–8251. https://doi.org/10.1109/tase.2024.3482362</mixed-citation><mixed-citation xml:lang="en">Tai Y., Yang K., Peng T., Huang Z., Zhang Z. Defect image sample generation with diffusion prior for steel surface defect recognition // IEEE Transactions on Automation Science and Engineering. 2024. V. 22. P. 8239–8251. https://doi.org/10.1109/tase.2024.3482362</mixed-citation></citation-alternatives></ref><ref id="cit6"><label>6</label><citation-alternatives><mixed-citation xml:lang="ru">Alzarooni A., Iqbal E., Ullah Khan, S. Javed S., Moyo B., Abdulrahman Y. Anomaly detection for industrial applications: challenges, solutions, and future directions // International Journal of Computer Integrated Manufacturing. 2025. https://doi.org/10.1080/0951192X.2025.2599548</mixed-citation><mixed-citation xml:lang="en">Alzarooni A., Iqbal E., Ullah Khan, S. Javed S., Moyo B., Abdulrahman Y. Anomaly detection for industrial applications: challenges, solutions, and future directions // International Journal of Computer Integrated Manufacturing. 2025. https://doi.org/10.1080/0951192X.2025.2599548</mixed-citation></citation-alternatives></ref><ref id="cit7"><label>7</label><citation-alternatives><mixed-citation xml:lang="ru">Bradski G. The openCV library // Dr. Dobb’s Journal: Software Tools for the Professional Programmer. 2000. V. 25. N 11. P. 120–123.</mixed-citation><mixed-citation xml:lang="en">Bradski G. The openCV library // Dr. Dobb’s Journal: Software Tools for the Professional Programmer. 2000. V. 25. N 11. P. 120–123.</mixed-citation></citation-alternatives></ref><ref id="cit8"><label>8</label><citation-alternatives><mixed-citation xml:lang="ru">Zhang Q., Li H. MOEA/D: A multiobjective evolutionary algorithm based on decomposition // IEEE Transactions on Evolutionary Computation. 2007. V. 11. N 6. P. 712–731. https://doi.org/10.1109/tevc.2007.892759</mixed-citation><mixed-citation xml:lang="en">Zhang Q., Li H. MOEA/D: A multiobjective evolutionary algorithm based on decomposition // IEEE Transactions on Evolutionary Computation. 2007. V. 11. N 6. P. 712–731. https://doi.org/10.1109/tevc.2007.892759</mixed-citation></citation-alternatives></ref><ref id="cit9"><label>9</label><citation-alternatives><mixed-citation xml:lang="ru">Rezatofighi H., Tsoi N., Gwak J., Sadeghian A., Reid I., Savarese S. Generalized intersection over union: A metric and a loss for bounding box regression // Proc. of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 2019. P. 658–666. https://doi.org/10.1109/cvpr.2019.00075</mixed-citation><mixed-citation xml:lang="en">Rezatofighi H., Tsoi N., Gwak J., Sadeghian A., Reid I., Savarese S. Generalized intersection over union: A metric and a loss for bounding box regression // Proc. of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 2019. P. 658–666. https://doi.org/10.1109/cvpr.2019.00075</mixed-citation></citation-alternatives></ref><ref id="cit10"><label>10</label><citation-alternatives><mixed-citation xml:lang="ru">Bakurov I., Buzzelli M., Schettini R., Castelli M., Vanneschi L. Structural similarity index (SSIM) revisited: a data-driven approach // Expert Systems with Applications. 2022. V. 189. P. 116087. https://doi.org/10.1016/j.eswa.2021.116087</mixed-citation><mixed-citation xml:lang="en">Bakurov I., Buzzelli M., Schettini R., Castelli M., Vanneschi L. Structural similarity index (SSIM) revisited: a data-driven approach // Expert Systems with Applications. 2022. V. 189. P. 116087. https://doi.org/10.1016/j.eswa.2021.116087</mixed-citation></citation-alternatives></ref><ref id="cit11"><label>11</label><citation-alternatives><mixed-citation xml:lang="ru">Zhang H., Liu Y., Yang J., Guo W., Wang X., Fua P. DiffAtlas: GenAIFying atlas segmentation via Image-Mask Diffusion // Lecture Notes in Computer Science. 2025. V. 15975. P. 161–172. https://doi.org/10.1007/978-3-032-05325-1_16</mixed-citation><mixed-citation xml:lang="en">Zhang H., Liu Y., Yang J., Guo W., Wang X., Fua P. DiffAtlas: GenAIFying atlas segmentation via Image-Mask Diffusion // Lecture Notes in Computer Science. 2025. V. 15975. P. 161–172. https://doi.org/10.1007/978-3-032-05325-1_16</mixed-citation></citation-alternatives></ref><ref id="cit12"><label>12</label><citation-alternatives><mixed-citation xml:lang="ru">Gatys L.A., Ecker A.S., Bethge M. Image style transfer using convolutional neural networks // Proc. of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2016. P. 2414– 2423. https://doi.org/10.1109/CVPR.2016.265</mixed-citation><mixed-citation xml:lang="en">Gatys L.A., Ecker A.S., Bethge M. Image style transfer using convolutional neural networks // Proc. of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2016. P. 2414– 2423. https://doi.org/10.1109/CVPR.2016.265</mixed-citation></citation-alternatives></ref><ref id="cit13"><label>13</label><citation-alternatives><mixed-citation xml:lang="ru">Jiang P., Ergu D., Liu F., Cai Y., Ma B. A Review of Yolo algorithm developments // Procedia Computer Science. 2022. V. 199. P. 1066– 1073. https://doi.org/10.1016/j.procs.2022.01.135</mixed-citation><mixed-citation xml:lang="en">Jiang P., Ergu D., Liu F., Cai Y., Ma B. A Review of Yolo algorithm developments // Procedia Computer Science. 2022. V. 199. P. 1066– 1073. https://doi.org/10.1016/j.procs.2022.01.135</mixed-citation></citation-alternatives></ref><ref id="cit14"><label>14</label><citation-alternatives><mixed-citation xml:lang="ru">Chen Y., Lin H., Zhang W., Chen W., Zhou Z., Heidari A.A., et al. ICycle-GAN: Improved cycle generative adversarial networks for liver medical image generation // Biomedical Signal Processing and Control. 2024. V. 92. P. 106100. https://doi.org/10.1016/j.bspc.2024.106100</mixed-citation><mixed-citation xml:lang="en">Chen Y., Lin H., Zhang W., Chen W., Zhou Z., Heidari A.A., et al. ICycle-GAN: Improved cycle generative adversarial networks for liver medical image generation // Biomedical Signal Processing and Control. 2024. V. 92. P. 106100. https://doi.org/10.1016/j.bspc.2024.106100</mixed-citation></citation-alternatives></ref><ref id="cit15"><label>15</label><citation-alternatives><mixed-citation xml:lang="ru">Qin D., Leichner C., Delakis M., Fornoni M., Luo S., Yang F., et al. MobileNetV4: universal models for the mobile ecosystem // Lecture Notes in Computer Science. 2025. V. 15098. P. 78–96. https://doi.org/10.1007/978-3-031-73661-2_5</mixed-citation><mixed-citation xml:lang="en">Qin D., Leichner C., Delakis M., Fornoni M., Luo S., Yang F., et al. MobileNetV4: universal models for the mobile ecosystem // Lecture Notes in Computer Science. 2025. V. 15098. P. 78–96. https://doi.org/10.1007/978-3-031-73661-2_5</mixed-citation></citation-alternatives></ref><ref id="cit16"><label>16</label><citation-alternatives><mixed-citation xml:lang="ru">Graham B., El-Nouby A., Touvron H., Stock P., Joulin A., Jégou H., et al. LeViT: a vision transformer in convnet’s clothing for faster inference // Proc. of the IEEE/CVF International Conference on Computer Vision (ICCV). 2021. P. 12239–12249. https://doi.org/10.1109/ICCV48922.2021.01204</mixed-citation><mixed-citation xml:lang="en">Graham B., El-Nouby A., Touvron H., Stock P., Joulin A., Jégou H., et al. LeViT: a vision transformer in convnet’s clothing for faster inference // Proc. of the IEEE/CVF International Conference on Computer Vision (ICCV). 2021. P. 12239–12249. https://doi.org/10.1109/ICCV48922.2021.01204</mixed-citation></citation-alternatives></ref><ref id="cit17"><label>17</label><citation-alternatives><mixed-citation xml:lang="ru">Ma N., Zhang X., Zheng H.-T., Sun J. ShuffleNet V2: practical guidelines for efficient CNN architecture design // Lecture Notes in Computer Science. 2018. V. 11218. P. 122–138. https://doi.org/10.1007/978-3-030-01264-9_8</mixed-citation><mixed-citation xml:lang="en">Ma N., Zhang X., Zheng H.-T., Sun J. ShuffleNet V2: practical guidelines for efficient CNN architecture design // Lecture Notes in Computer Science. 2018. V. 11218. P. 122–138. https://doi.org/10.1007/978-3-030-01264-9_8</mixed-citation></citation-alternatives></ref><ref id="cit18"><label>18</label><citation-alternatives><mixed-citation xml:lang="ru">Cai H., Li J., Hu M., Gan C., Han S. EfficientViT: lightweight multiscale attention for high-resolution dense prediction // Proc. of the IEEE/CVF International Conference on Computer Vision (ICCV). 2023 . P. 17256 – 17267. https://doi.org/10.1109/ICCV51070.2023.01587</mixed-citation><mixed-citation xml:lang="en">Cai H., Li J., Hu M., Gan C., Han S. EfficientViT: lightweight multiscale attention for high-resolution dense prediction // Proc. of the IEEE/CVF International Conference on Computer Vision (ICCV). 2023 . P. 17256 – 17267. https://doi.org/10.1109/ICCV51070.2023.01587</mixed-citation></citation-alternatives></ref><ref id="cit19"><label>19</label><citation-alternatives><mixed-citation xml:lang="ru">Khanam R., Hussain M. YOLOv11: an overview of the key architectural enhancements // arXiv. 2024. arXiv:2410.17725. https://doi.org/10.48550/arXiv.2410.17725</mixed-citation><mixed-citation xml:lang="en">Khanam R., Hussain M. YOLOv11: an overview of the key architectural enhancements // arXiv. 2024. arXiv:2410.17725. https://doi.org/10.48550/arXiv.2410.17725</mixed-citation></citation-alternatives></ref><ref id="cit20"><label>20</label><citation-alternatives><mixed-citation xml:lang="ru">Hara K., Kataoka H., Satoh Y. Learning spatio-temporal features with 3D residual networks for action recognition // Proc. of the IEEE International Conference on Computer Vision Workshops (ICCVW). 2017. P. 3154–3160. https://doi.org/10.1109/iccvw.2017.373</mixed-citation><mixed-citation xml:lang="en">Hara K., Kataoka H., Satoh Y. Learning spatio-temporal features with 3D residual networks for action recognition // Proc. of the IEEE International Conference on Computer Vision Workshops (ICCVW). 2017. P. 3154–3160. https://doi.org/10.1109/iccvw.2017.373</mixed-citation></citation-alternatives></ref><ref id="cit21"><label>21</label><citation-alternatives><mixed-citation xml:lang="ru">Imambi S., Prakash K.B., Kanagachidambaresan G.R. PyTorch // Programming with TensorFlow: Solution for Edge Computing Applications. 2021. P. 87–104. https://doi.org/10.1007/978-3-030-57077-4_10</mixed-citation><mixed-citation xml:lang="en">Imambi S., Prakash K.B., Kanagachidambaresan G.R. PyTorch // Programming with TensorFlow: Solution for Edge Computing Applications. 2021. P. 87–104. https://doi.org/10.1007/978-3-030-57077-4_10</mixed-citation></citation-alternatives></ref><ref id="cit22"><label>22</label><citation-alternatives><mixed-citation xml:lang="ru">Loshchilov I., Hutter F. Fixing weight decay regularization in Adam // arXiv. 2017.arXiv:1711.05101v2.</mixed-citation><mixed-citation xml:lang="en">Loshchilov I., Hutter F. Fixing weight decay regularization in Adam // arXiv. 2017.arXiv:1711.05101v2.</mixed-citation></citation-alternatives></ref><ref id="cit23"><label>23</label><citation-alternatives><mixed-citation xml:lang="ru">Meyer G.P. An alternative probabilistic interpretation of the Huber loss // Proc. of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 2021. P. 5257–5265. https://doi.org/10.1109/cvpr46437.2021.00522</mixed-citation><mixed-citation xml:lang="en">Meyer G.P. An alternative probabilistic interpretation of the Huber loss // Proc. of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 2021. P. 5257–5265. https://doi.org/10.1109/cvpr46437.2021.00522</mixed-citation></citation-alternatives></ref><ref id="cit24"><label>24</label><citation-alternatives><mixed-citation xml:lang="ru">Huang P., Tian S., Su Y., Tan W., Dong Y., Xu W., et al. IA-CIOU: an improved IOU bounding box loss function for SAR ship target detection methods // IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing. 2024. V. 17. P. 10569–10582. https://doi.org/10.1109/jstars.2024.3402540</mixed-citation><mixed-citation xml:lang="en">Huang P., Tian S., Su Y., Tan W., Dong Y., Xu W., et al. IA-CIOU: an improved IOU bounding box loss function for SAR ship target detection methods // IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing. 2024. V. 17. P. 10569–10582. https://doi.org/10.1109/jstars.2024.3402540</mixed-citation></citation-alternatives></ref><ref id="cit25"><label>25</label><citation-alternatives><mixed-citation xml:lang="ru">Wan E.A., Van Der Merwe R. The unscented Kalman filter for nonlinear estimation // Proc. of the IEEE 2000 Adaptive Systems for Signal Processing, Communications, and Control Symposium (Cat. No.00EX373). 2000. P. 153–158. https://doi.org/10.1109/ASSPCC.2000.882463</mixed-citation><mixed-citation xml:lang="en">Wan E.A., Van Der Merwe R. The unscented Kalman filter for nonlinear estimation // Proc. of the IEEE 2000 Adaptive Systems for Signal Processing, Communications, and Control Symposium (Cat. No.00EX373). 2000. P. 153–158. https://doi.org/10.1109/ASSPCC.2000.882463</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>
