Журнал

About Editorial Board Information for authors and reviewers Publication ethics Contacts User Account: send / review a manuscript
Archive
Volume 22, 2025
Number 1 Number 2 Number 3 Number 4 Number 5
Volume 21, 2024
Number 1 Number 2 Number 3 Number 4 Number 5 Number 6
Volume 20, 2023
Number 1 Number 2 Number 3 Number 4 Number 5 Number 6
Volume 19, 2022
Number 1 Number 2 Number 3 Number 4 Number 5 Number 6
Volume 18, 2021
Number 1 Number 2 Number 3 Number 4 Number 5 Number 6
Volume 17, 2020
Number 1 Number 2 Number 3 Number 4 Number 5 Number 6 Number 7
Volume 16, 2019
Number 1 Number 2 Number 3 Number 4 Number 5 Number 6
Volume 15, 2018
Number 1 Number 2 Number 3 Number 4 Number 5 Number 6 Number 7
Volume 14, 2017
Number 1 Number 2 Number 3 Number 4 Number 5 Number 6 Number 7
Volume 13, 2016
Number 1 Number 2 Number 3 Number 4 Number 5 Number 6
Volume 12, 2015
Number 1 Number 2 Number 3 Number 4 Number 5 Number 6
Volume 11, 2014
Number 1 Number 2 Number 3 Number 4
Volume 10, 2013
Number 1 Number 2 Number 3 Number 4
Volume 9, 2012
Number 1 Number 2 Number 3 Number 4 Number 5
Volume 8, 2011
Number 1 Number 2 Number 3 Number 4
Volume 7, 2010
Number 1 Number 2 Number 3 Number 4
Issue 6, 2009
Volume 1 Volume 2
Issue 5, 2008
Volume 1 Volume 2
Issue 4, 2007
Volume 1 Volume 2
Issue 3, 2006
Volume 1 Volume 2
Issue 2, 2005
Volume 1 Volume 2
Issue 1, 2004
Volume 1
Search
Find:
русский
ISSN 2070-7401 (Print), ISSN 2411-0280 (Online)
Sovremennye problemy distantsionnogo zondirovaniya Zemli iz kosmosa
CURRENT PROBLEMS IN REMOTE SENSING OF THE EARTH FROM SPACE

  

Sovremennye problemy distantsionnogo zondirovaniya Zemli iz kosmosa, 2025, V. 22, No. 5, pp. 51-62

Segmentation of small objects in radar images using a graph-convolutional neural network implementation of quadtrees

A.M. Dostovalova 1 , A.K. Gorshenin 1 
1 Federal Research Center “Computer Science and Control” RAS, Moscow, Russia
Accepted: 07.08.2025
DOI: 10.21046/2070-7401-2025-22-5-51-62
The paper proposes an architecture for a balanced neural network quadtree to solve the problem of segmenting small objects in radar images with insufficient training data. This architecture includes a pretrained convolutional encoder that generates additional features for each pixel in the image, as well as a graph-convolutional neural network that enhances the processing of spatial relationships in the data through an integrated probabilistic quadtree model. The graph part contains a branch-pruning block that detects similarities between pixels across different spatial resolutions. Additionally, there is a feature-processing unit designed for scaled images, and a custom loss function to improve the accuracy of separating unbalanced classes. The U-Net architecture was used for segmenting radar images from Sentinel-1 and High Resolution Synthetic Aperture Radar Images Dataset (HRSID). The balanced neural network quadtree demonstrated improved segmentation quality for smaller objects in both multi-class and binary segmentation (i.e., object detection against the background) compared to a basic neural network quadtree and vanilla U-Net. As a result, F1 scores for small object classes increased by 3.59% for binary segmentation and by 47.42% for multi-class segmentation. The developed approaches have potential to be applied to the problem of detection of small-size objects using superpixels on high-resolution images.
Keywords: neural quadtrees, probability-informed neural networks, graph convolution networks, SAR image segmentation, small objects
Full text

References:

  1. Dostovalova A. M., Neural quadtree and its applications for SAR imagery segmentation, Informatika i ee primeneniya, 2024, V. 18, Iss. 3, P. 77–85 (in Russian), DOI: 10.14357/19922264240410.
  2. Bergstra J., Bengio Y., Random search for hyper-parameter optimization, J. Machine Learning Research, 2012, V. 13, Iss. 10, pp. 281–305.
  3. Chitta K., Álvarez J. M., Hebert M., Quadtree generating networks: Efficient hierarchical scene parsing with sparse convolutions, Proc. 2020 IEEE Winter Conf. Applications of Computer Vision (WACV), Snowmass, CO, USA, 2020, pp. 2009–2018, DOI: 10.1109/WACV45572.2020.9093449.
  4. Dostovalova A. M., Using a model of a spatial–hierarchical quadtree with truncated branches to improve the accuracy of image classification, Izvestiya, Atmospheric and Oceanic Physics, 2023, V. 59, Iss. 12, pp. 1255–1262, DOI: 10.1134/S0001433823120071.
  5. Dostovalova A. M., Gorshenin A. K., Neural network image classifiers informed by factor analyzers, Doklady Mathematics, 2024, V. 110, pp. S35–S41, DOI: 10.1134/S106456242460204X.
  6. Dostovalova A., Gorshenin A., Small sample learning based on probability-informed neural networks for SAR image segmentation, Neural Computing and Applications, 2025, V. 35, pp. 8285–8308, DOI: 10.1007/s00521-025-10997-x.
  7. Friedman M., A comparison of alternative tests of significance for the problem of m rankings, The Annals of Mathematical Statistics, 1940, V. 11, Iss. 1, pp. 86–92, DOI: 10.1214/aoms/1177731944.
  8. Gorshenin A. K., Kuzmin V. Yu., Statistical feature construction for forecasting accuracy increase and its applications in neural network based analysis, Mathematics, 2022, V. 10, Article 589, DOI: 10.3390/math10040589.
  9. Gorshenin A. K., Vilyaev A. L., Finite normal mixture models for the ensemble learning of recurrent neural networks with applications to currency pairs, Pattern Recognition and Image Analysis, 2022, Iss. 32, pp. 780–792, DOI: 10.1134/S1054661822040058.
  10. Gorshenin A. K., Vilyaev A. L., Machine learning models informed by connected mixture components for short- and medium-term time series forecasting, AI, 2024, V. 5, Iss. 4, pp. 1955–1976, DOI: 10.3390/ai5040097.
  11. Gorshenin A., Kozlovskaya A., Gorbunov S., Kochetkova I., Mobile network traffic analysis based on probability-informed machine learning approach, Computer Networks, 2024, V. 247, Article 110433, DOI: 10.1016/j.comnet.2024.110433.
  12. Jayaraman P. K., Mei J., Cai J., Zheng J., Quadtree convolutional neural networks, ECCV 2018. Lecture Notes in Computer Science, 2018, V. 11210, pp. 554–569, DOI: 10.1007/978-3-030-01231-1_34.
  13. Jewsbury R., Bhalerao A., Rajpoot N. M., A quadtree image representation for computational pathology, Proc. 2021 IEEE/CVF Intern. Conf. on Computer Vision Workshops (ICCVW), Montreal, BC, Canada, 2021, pp. 648–656, DOI: 10.1109/ICCVW54120.2021.00078.
  14. Karniadakis G. E., Kevrekidis I. G., Lu L. et al., Physics-informed machine learning, Nature Reviews Physics, 2021, V. 3, pp. 422–440, DOI: 10.1038/s42254-021-00314-5.
  15. Ke L., Danelljan M., Li X. et al., Mask Transfiner for high-quality instance segmentation, Proc. 2022 IEEE/CVF Conf. Computer Vision and Pattern Recognition (CVPR), New Orleans, LA, USA, 2022, pp. 4402–4411, DOI: 10.1109/CVPR52688.2022.00437.
  16. Kochetkova I., Kushchazli A., Burtseva S., Gorshenin A., Short-term mobile network traffic forecasting using seasonal ARIMA and Holt-Winters models, Future Internet, 2023, V. 15, Iss. 9, Article 290, DOI: 10.3390/fi15090290.
  17. Ma F., Gao F., Sun J. et al., Attention Graph Convolution Network for image segmentation in big SAR imagery data, Remote Sensing, 2019, V. 11, Iss. 21, Article 2586, DOI: 10.3390/rs11212586.
  18. Özdemir Ö., Sönmez E. B., Weighted Cross-Entropy for unbalanced data with application on COVID X-ray images, 2020 Innovations in Intelligent Systems and Applications Conference (ASYU), Istanbul, Turkey, 2020, pp. 1–6, DOI: 10.1109/ASYU50717.2020.9259848.
  19. Pastorino M., Montaldo A., Fronda L. et al., Multisensor and multiresolution remote sensing image classification through a causal hierarchical Markov framework and decision tree ensembles, Remote Sensing, 2021, V. 13, Iss. 5, Article 849, DOI: 10.3390/rs13050849.
  20. Pastorino M., Moser G., Serpico S. B., Zerubia J., Semantic segmentation of remote-sensing images through fully convolutional neural networks and hierarchical probabilistic graphical models, IEEE Trans. Geoscience and Remote Sensing, 2022, V. 60, pp. 1–16, DOI: 10.1109/TGRS.2022.3141996.
  21. Potin P., Bargellini P., Laur H. et al., Sentinel-1 mission operations concept, Proc. IEEE Intern. Geoscience and Remote Sensing Symp., Munich, Germany, 2012, pp. 1745–1749.
  22. Powers D. M. W., Evaluation: from precision, recall and F-measure to ROC, informedness, markedness and correlation, Intern. J. Machine Learning Technology, 2011, V. 2, Iss. 1, pp. 37–63.
  23. Ronen T., Levy O., Golbert A., Vision transformers with mixed-resolution tokenization, Proc. 2023 IEEE/CVF Conf. Computer Vision and Pattern Recognition (CVPRW), Vancouver, BC, Canada, 2023, pp. 4613–4622, DOI: 10.1109/CVPRW59228.2023.00486.
  24. Ronneberger O., Fischer P., Brox T., U-Net: Convolutional networks for biomedical image segmentation, MICCAI 2015. Lecture Notes in Computer Science, 2015, V. 9351, pp. 234–241, DOI: 10.1007/978-3-319-24574-4_28.
  25. Sasmal B., Dhal K. G., A survey on the utilization of Superpixel image for clustering based image segmentation, Multimedia Tools and Applications, 2023, V. 82, pp. 35493–35555, DOI: 10.1007/s11042-023-14861-9.
  26. Tang S., Zhang J., Zhu S., Tan P., Quadtree attention for vision transformers, https://arxiv.org/, arXiv:2201.02767, 2022, 16 p., DOI: 10.48550/arXiv.2201.02767.
  27. Wei Sh., Zeng X., Que Q. et al., HRSID: A high-resolution SAR images dataset for ship detection and instance segmentation, IEEE Access, 2020, V. 8, pp. 120234–120254, DOI: 10.1109/ACCESS.2020.3005861.