informИнформатикаInformatics1816-03012617-6963UIIP NASB10.37661/1816-0301-2025-22-2-33-47inform-1351Research ArticleОБРАБОТКА СИГНАЛОВ, ИЗОБРАЖЕНИЙ, РЕЧИ, ТЕКСТА И РАСПОЗНАВАНИЕ ОБРАЗОВSIGNAL, IMAGE, SPEECH, TEXT PROCESSING AND PATTERN RECOGNITIONЭффективное обнаружение зданий на изображениях дистанционного зондирования на основе улучшенной сети YOLOv10Efficient detection of building in remote sensing images using an improved YOLOv10 networkhttps://orcid.org/0009-0003-6976-5386ВуС.WuX.

Ву Сяньи, аспирант механико-математического факультета

пр. Независимости, 4, Минск, 220030

Xianyi Wu, Postgraduate Student of the Faculty of Mechanics and Mathematics

av. Nezavisimosti, 4, Minsk, 220030

tigerv5872@gmail.comАбламейкоС. В.AblameykoSe. V.

Абламейко Сергей Владимирович, академик НАН Беларуси, доктор технических наук, профессор механико-математического факультета

ул. Сурганова, 6, Минск, 220012

Sergey V. Ablameyko, Acad. of the National Academy of Sciences of Belarus, D. Sc. (Eng.), Prof. of the Faculty of Mechanics and Mathematics

st. Surganova, 6, Minsk, 220012

ablameyko@bsu.byБелорусский государственный университетBelarusian State UniversityОбъединенный институт проблем информатики Национальной академии наукThe United Institute of Informatics Problems of the National Academy of Sciences of Belarus2025100720252223347Copyright © Ву С., Абламейко С.В., 20252025Ву С., Абламейко С.В.Wu X., Ablameyko S.V.Данная работа распространяется под лицензией Creative Commons Attribution 4.0.This work is licensed under a Creative Commons Attribution 4.0 License.https://inf.grid.by/jour/article/view/1351

Цели . В настоящее время быстрое определение местоположения и размера объектов зданий с помощью изображений дистанционного зондирования имеет важное научно-исследовательское и практическое значение для городского планирования, мониторинга окружающей среды и управления стихийными бедствиями.Методы . Предлагается метод обнаружения объектов на основе улучшенной сети YOLOv10, которая включает в себя механизм внимания Супертокен, модель RepConv (повторно параметризуемая свертка) и нормализованное взвешенное расстояние для более точного обнаружения зданий на изображениях дистанционного зондирования. Метод повышает точность и эффективность обнаружения, особенно для небольших объектов. Набор данных LEVIR-CD используется для обучения и тестирования модели.Результаты . Экспериментальные результаты показывают, что предлагаемый метод демонстрирует лучшую точность при решении задачи обнаружения зданий, чем традиционный YOLOv10 и другие методы.Заключение. Предлагаемый метод эффективно повышает точность и эффективность обнаружения зданий на изображениях дистанционного зондирования

Objectives. At present, rapid detection of the location and size of building objects from remote sensing images is important for scientific research value and has practical significance for urban planning, environmental monitoring and disaster management.Methods. This paper proposes an object detection method based on improved YOLOv10 network, which incorporates Super Token Attention, RepConv and Normalized Weighted Distance to more precisely detect buildings in remote sensing images. This method improves the detection accuracy and efficiency especially for small objects. The LEVIR-CD dataset is used for model training and testing.Results. The experimental results show that the method demonstrates better accuracy on the building detection task than the traditional YOLOv10 and other methods.Conclusion. The proposed method significantly enhances the accuracy and efficiency of building detection in remote sensing images

YOLOv10изображения дистанционного зондированиямеханизм вниманияобнаружение зданийRepConvSuper Token AttentionYOLOv10remote sensing imagesattention mechanismbuilding detectionRepConvSuper Token AttentionReferencesLi S. T., Li C. Y., Kang X. D. Current status and future prospects of multi-source remote sensing image fusion. National Remote Sensing Bulletin, 2021, vol. 25, no. 1, pp. 148–166. DOI: 10.11834/jrs.20210259.Li S. T., Li C. Y., Kang X. D. Current status and future prospects of multi-source remote sensing image fusion. National Remote Sensing Bulletin, 2021, vol. 25, no. 1, pp. 148–166. DOI: 10.11834/jrs.20210259.Luo H. L., Wang W. X., Ye X. Y., Zhu S. X., Bai Y. Q. Research progress on directed object detection based on deep learning. Image and Signal Processing, 2024, vol. 13, no. 3, pp. 258–270. DOI: 10.12677/jisp.2024.133022.Luo H. L., Wang W. X., Ye X. Y., Zhu S. X., Bai Y. Q. Research progress on directed object detection based on deep learning. Image and Signal Processing, 2024, vol. 13, no. 3, pp. 258–270. DOI: 10.12677/jisp.2024.133022.Abdikan S., Bilgin G., Sanli F. B., Uslu E., Ustuner M. Enhancing land use classification with fusing dualpolarized terrasar-x and multispectral rapideye data. Journal of Applied Remote Sensing, 2015, vol. 9, no. 1, p. 096054. DOI: 10.1117/1.JRS.9.096054.Abdikan S., Bilgin G., Sanli F. B., Uslu E., Ustuner M. Enhancing land use classification with fusing dualpolarized terrasar-x and multispectral rapideye data. Journal of Applied Remote Sensing, 2015, vol. 9, no. 1, p. 096054. DOI: 10.1117/1.JRS.9.096054.Liu F. F., Zhu C. M., Zhao N. N., Wu J. H. Remote sensing small target detection based on multimodal fusion. Laser & Optoelectronics Progress, 2024, vol. 61, no. 24, p. 2428010. DOI: 10.3788/LOP241203.Liu F. F., Zhu C. M., Zhao N. N., Wu J. H. Remote sensing small target detection based on multimodal fusion. Laser & Optoelectronics Progress, 2024, vol. 61, no. 24, p. 2428010. DOI: 10.3788/LOP241203.Li J., Wei X. M. Research on efficient detection network method for remote sensing images based on selfattention mechanism. Image and Vision Computing, 2024, vol. 142, p. 104884. DOI: 10.1016/j.imavis.2023.104884.Li J., Wei X. M. Research on efficient detection network method for remote sensing images based on selfattention mechanism. Image and Vision Computing, 2024, vol. 142, p. 104884. DOI: 10.1016/j.imavis.2023.104884.Liu D., Zhong L., Wu H., Li S., Li Y. Remote sensing image super-resolution reconstruction by fusing multi-scale receptive fields and hybrid transformer. Scientific Reports, 2025, vol. 15, p. 2140. DOI: 10.1038/s41598-025-86446-5.Liu D., Zhong L., Wu H., Li S., Li Y. Remote sensing image super-resolution reconstruction by fusing multi-scale receptive fields and hybrid transformer. Scientific Reports, 2025, vol. 15, p. 2140. DOI: 10.1038/s41598-025-86446-5.Li Z., Wang H., Ma G., Yang W., Ablameyko S. Effective small object detection in remote sensing images based on improved YOLOv8 network. Nonlinear Phenomena in Complex Systems, 2024, vol. 27, no. 3, pp. 278–291. DOI: 10.5281/zenodo.13960639.Li Z., Wang H., Ma G., Yang W., Ablameyko S. Effective small object detection in remote sensing images based on improved YOLOv8 network. Nonlinear Phenomena in Complex Systems, 2024, vol. 27, no. 3, pp. 278–291. DOI: 10.5281/zenodo.13960639.Tan M., Le Q. V. EfficientNetV2: Smaller models and faster training, 2021. Available at: https://arxiv.org/abs/2104.00298 (accessed 13.02.2025). DOI: 10.48550/arXiv.2104.00298. (Preprint).Tan M., Le Q. V. EfficientNetV2: Smaller models and faster training, 2021. Available at: https://arxiv.org/abs/2104.00298 (accessed 13.02.2025). DOI: 10.48550/arXiv.2104.00298. (Preprint).Woo S., Park J., Lee J. Y., Kweon I. S. CBAM: Convolutional Block Attention Module. Proceedings of the European Conference on Computer Vision (ECCV), Munich, Germany, 8–14 September 2018. Springer, Cham, 2018, pp. 3–19. DOI: 10.1007/978-3-030-01234-2_1.Woo S., Park J., Lee J. Y., Kweon I. S. CBAM: Convolutional Block Attention Module. Proceedings of the European Conference on Computer Vision (ECCV), Munich, Germany, 8–14 September 2018. Springer, Cham, 2018, pp. 3–19. DOI: 10.1007/978-3-030-01234-2_1.Wang A., Chen H., Lin Z., Han J., Ding G. RepViT: Revisiting mobile CNN from ViT perspective, 2023. Available at: https://arxiv.org/abs/2307.09283 (accessed 13.02.2025). DOI: 10.48550/arXiv.2307.09283. (Preprint).Wang A., Chen H., Lin Z., Han J., Ding G. RepViT: Revisiting mobile CNN from ViT perspective, 2023. Available at: https://arxiv.org/abs/2307.09283 (accessed 13.02.2025). DOI: 10.48550/arXiv.2307.09283. (Preprint).Dosovitskiy A., Beyer L., Kolesnikov A., Weissenborn D., Zhai X., …, Houlsby N. An image is worth 16×16 words: Transformers for image recognition at scale, 2020. Available at: https://arxiv.org/abs/2010.11929 (accessed 13.02.2025). DOI: 10.48550/arXiv.2010.11929. (Preprint).Dosovitskiy A., Beyer L., Kolesnikov A., Weissenborn D., Zhai X., …, Houlsby N. An image is worth 16×16 words: Transformers for image recognition at scale, 2020. Available at: https://arxiv.org/abs/2010.11929 (accessed 13.02.2025). DOI: 10.48550/arXiv.2010.11929. (Preprint).Wang X., Zhu D., Yan Y. Towards efficient detection for small objects via attention-guided detection network and data augmentation. Sensors, 2022, vol. 22, no. 19, p. 7663. DOI: 10.3390/s22197663.Wang X., Zhu D., Yan Y. Towards efficient detection for small objects via attention-guided detection network and data augmentation. Sensors, 2022, vol. 22, no. 19, p. 7663. DOI: 10.3390/s22197663.Wang A., Chen H., Liu L. H., Chen K., Lin Z. J., …, Ding G. G. YOLOv10: Real-time end-to-end object detection, 2024. Available at: https://doi.org/10.48550/arXiv.2405.14458 (accessed 13.02.2025). DOI: 10.48550/arXiv.2405.14458. (Preprint).Wang A., Chen H., Liu L. H., Chen K., Lin Z. J., …, Ding G. G. YOLOv10: Real-time end-to-end object detection, 2024. Available at: https://doi.org/10.48550/arXiv.2405.14458 (accessed 13.02.2025). DOI: 10.48550/arXiv.2405.14458. (Preprint).Wang J., Xu C., Yang W., Yu L. A normalized Gaussian Wasserstein distance for tiny object detection, 2021. Available at: https://doi.org/10.48550/arXiv.2110.13389 (accessed 13.02.2025). DOI: 10.48550/arXiv.2110.13389. (Preprint).Wang J., Xu C., Yang W., Yu L. A normalized Gaussian Wasserstein distance for tiny object detection, 2021. Available at: https://doi.org/10.48550/arXiv.2110.13389 (accessed 13.02.2025). DOI: 10.48550/arXiv.2110.13389. (Preprint).Huang H. B., Zhou X. Q., Cao J., He R., Tan T. N. Vision transformer with super token sampling, 2022. Available at: https://doi.org/10.48550/arXiv.2211.11167 (accessed 13.02.2025). DOI: 10.48550/arXiv.2211.11167. (Preprint).Huang H. B., Zhou X. Q., Cao J., He R., Tan T. N. Vision transformer with super token sampling, 2022. Available at: https://doi.org/10.48550/arXiv.2211.11167 (accessed 13.02.2025). DOI: 10.48550/arXiv.2211.11167. (Preprint).Wan D. H., Lu R., Tian S., Xu T., Lang X., Ren Z. Mixed local channel attention for object detection. Engineering Applications of Artificial Intelligence, 2023, vol. 123, p. 106442. DOI: 10.1016/j.engappai.2023.106442.Wan D. H., Lu R., Tian S., Xu T., Lang X., Ren Z. Mixed local channel attention for object detection. Engineering Applications of Artificial Intelligence, 2023, vol. 123, p. 106442. DOI: 10.1016/j.engappai.2023.106442.Wang H., Ablameyko S. Enhancing small object detection in remote sensing images using mixed local channel attention with YOLOv8. Journal of Computer Technology and Applied Mathematics, 2024, vol. 1, no. 1, pp. 40–45. DOI: 10.5281/zenodo.10986298.Wang H., Ablameyko S. Enhancing small object detection in remote sensing images using mixed local channel attention with YOLOv8. Journal of Computer Technology and Applied Mathematics, 2024, vol. 1, no. 1, pp. 40–45. DOI: 10.5281/zenodo.10986298.Ding X., Zhang X., Ma N., Han J., Ding G., Sun J. RepVGG: Making VGG-style ConvNets great again. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Nashville, TN, USA, 20–25 June 2021, рр. 13 733–13 742.Ding X., Zhang X., Ma N., Han J., Ding G., Sun J. RepVGG: Making VGG-style ConvNets great again. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Nashville, TN, USA, 20–25 June 2021, рр. 13 733–13 742.Vasa V. K., Zhu W., Chen X., Qiu P., Dong X., Wang Y. STA-Unet: Rethink the semantic redundant for medical imaging segmentation, 2024. Available at: https://arxiv.org/pdf/2410.11578 (accessed 13.02.2025). DOI: 10.48550/arXiv.2410.11578. (Preprint).Vasa V. K., Zhu W., Chen X., Qiu P., Dong X., Wang Y. STA-Unet: Rethink the semantic redundant for medical imaging segmentation, 2024. Available at: https://arxiv.org/pdf/2410.11578 (accessed 13.02.2025). DOI: 10.48550/arXiv.2410.11578. (Preprint).Yu Z., Huang H., Chen W., Su Y., Liu Y., Wang X. YOLO-FaceV2: A scale and occlusion aware face detector, 2022. Available at: https://arxiv.org/pdf/2208.02019v2 (accessed 13.02.2025). DOI: 10.48550/arXiv.2208.02019. (Preprint).Yu Z., Huang H., Chen W., Su Y., Liu Y., Wang X. YOLO-FaceV2: A scale and occlusion aware face detector, 2022. Available at: https://arxiv.org/pdf/2208.02019v2 (accessed 13.02.2025). DOI: 10.48550/arXiv.2208.02019. (Preprint).Chen H., Shi Z. A spatial-temporal attention-based method and a new dataset for remote sensing image change detection. Remote Sensing, 2020, vol. 12, no. 10, p. 1662. DOI: 10.3390/rs12101662.Chen H., Shi Z. A spatial-temporal attention-based method and a new dataset for remote sensing image change detection. Remote Sensing, 2020, vol. 12, no. 10, p. 1662. DOI: 10.3390/rs12101662.

The authors declare that there are no conflicts of interest present.