Using the Mask R-CNN model for segmentation of real estate objects in aerial photographs

Cover Page

Cite item

Full Text

Abstract

The mass appearance of illegal and unregistered in the Unified State Register of Real Estate (USRRE) real estate objects complicates cadastral registration for many entities at the territorial and administrative levels. Traditional methods of identifying objects of this type, based on manual analysis of geospatial data, are labor-intensive and time-consuming.To improve the efficiency of this process, it is proposed to automate the detection of objects in aerial photographs by solving the instance segmentation problem using the Mask R-CNN deep learning model. The article describes the preparation of a dataset for this model, examines the main quality metrics, and analyzes the results obtained. The efficiency of the Mask R-CNN model in practice is shown for solving the problem of detecting construction projects that are not registered in the USRRE.

About the authors

Igor Victorovich Vinokurov

Financial University under the Government of the Russian Federation

Email: igvvinokurov@fa.ru
Candidate of Technical Sciences (PhD), Associate Professor at the Financial University under the Government of the Russian Federation. Research interests: information systems, information technologies, data processing technologies

References

  1. T.-Y. Lin, P. Goyal, R. B. Girshick, K. He, P. Dollár. Focal loss for dense object detection, Computing Research Repository (CoRR), 2017, 10 pp.
  2. X. Wang, T. Kong, Ch. Shen, Y. Jiang, L. Li. SOLO: Segmenting objects by locations, Computing Research Repository (CoRR), 2019, 19 pp.
  3. K. Duda, A. Ivanov. „On decidability of amenability in computable groups“, Archive for Mathematical Logic, 61 (2022), pp. 891–902.
  4. K. He, G. Gkioxari, P. Dollár, R. B. Girshick. Mask R-CNN, Computing Research Repository (CoRR), 2017, 12 pp.
  5. S. Ren, K. He, R. B. Girshick, J. Sun. Faster R-CNN: Towards real-time object detection with region proposal networks, Computing Research Repository (CoRR), 2015, 14 pp.
  6. K. He, X. Zhang, S. Ren, J. Sun. Identity mappings in deep residual networks, Computing Research Repository (CoRR), 2016, 15 pp.
  7. T.-Y. Lin, M. Maire, S. J. Belongie, L. D. Bourdev, R. B. Girshick, J. Hays, P. Perona, D. Ramanan, P. Dollár, C. L. Zitnick. Microsoft COCO: Common objects in context, Computing Research Repository (CoRR), 2014, 15 pp.
  8. Y. Xu, L. Wu, Z. Xie, Z. Chen. „Building extraction in very high resolution remote sensing imagery using deep learning and guided filters“, Remote. Sens., 10:1 (2018), 144, 18 pp.
  9. Q. Han, Q. Yin, X. Zheng, Z. Chen. „Remote sensing image building detection method based on Mask R-CNN“, Complex Intell. Syst., 8 (2022), pp. 1847—1855.
  10. K. Zhao, J. Kang, J. Jung, G. Sohn. „Building extraction from satellite images using Mask R-CNN with building boundary regularization“, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) (18–22 June 2018, Salt Lake City, UT, USA), IEEE, 4 pp.
  11. X. Nie, M. Duan, H. Ding, B. Hu, E. K. Wong. „Attention Mask R-CNN for ship detection and segmentation from remote sensing images“, IEEE Access, 8 (2020), pp. 9325–9334.
  12. M. Jenila Vincent, P. Varalakshmi. „Extraction of building footprint using MASK-RCNN for high resolution aerial imagery“, Environmental Research Communications, 6:7 (2024), 075015, 17 pp.
  13. X. Zhu, L. Hu, J. Wang. „Urban modern architecture recognition based on Mask-RCNN and ECA attention mechanism“, Fifth International Conference on Geoscience and Remote Sensing Mapping (ICGRSM 2023) (13–15 October 2023, Lianyungang, China), Proc. SPIE, vol. 12980, 2024, ISBN 9781510672789.
  14. R. Raghavan, D. Chander Verma, D. Pandey, R. Anand, B. Kumar Pandey, H. Singh. „Optimized building extraction from high-resolution satellite imagery using deep learning“, Multimedia Tools and Applications, 81:29 (2022), pp. 42309–42323.
  15. D. Ulanov, A. Syrov. „Building footprint extraction based on RGBD satellite imagery“, CS230 Deep Learning (Winter 2020, Stanford University, CA), 2020, 11 pp.
  16. A. Solanki, R. K. Singh, B. Demeneze. „Aerial pictures semantic segmentation applying deep learning“, International Journal of Trendy Research in Engineering and Technology, 5:1 (2021), pp. 42–48.
  17. A. NourEldeen, M. E. Wahed. „Enhanced building footprint extraction from satellite imagery using Mask R-CNN and PointRend“, Bulletin of Electrical Engineering and Informatics, 5:13 (2024), pp. 3601–3608.
  18. K. He, X. Zhang, S. Ren, J. Sun. Deep residual learning for image recognition, Computing Research Repository (CoRR), 2015, 12 pp.
  19. Ch. J. Mills. PyTorch Mask R-CNN tutorial, GitHub repository, 2023 URL https://github.com/cj-mills/pytorch-mask-rcnn-tutorial-code.
  20. J. Redmon, S. Divvala, R. B. Girshick, A. Farhadi. You Only Look Once: Unified, real-time object detection, Computing Research Repository (CoRR), 2015, 10 pp.
  21. R. Khanam, M. Hussain. YOLOv11: An overview of the key architectural enhancements, 2024, 9 pp.
  22. T.-Y. Lin, P. Dollár, R. B. Girshick, K. He, B. Hariharan, S. J. Belongie. Feature pyramid networks for object detection, Computing Research Repository (CoRR), 2016, 10 pp.
  23. A. Waleed. Mask R-CNN for object detection and instance segmentation on Keras and TensorFlow, GitHub repository, 2017 URL https://github.com/matterport/Mask_RCNN.
  24. E. Stevens, L. Antiga, T. Viehmann. Deep Learning with PyTorch, Manning Publications, New York, 2020, ISBN 9781617295263, 520 pp.
  25. R. Sapkota, A. Dawood, M. Karkee. „Comparing YOLOv8 and Mask R-CNN for instance segmentation in complex orchard environments“, Artificial Intelligence in Agriculture, 13:1 (2024), pp. 84–99.
  26. V. F. Bulavitsky. „Use of drones to obtain aerial photographas terrain“, Elektronnoe nauchnoe izdanie «Uchyonye zametki TOGU», 4:4 (2013), pp. 1747–1755 (in Russian).

Supplementary files

Supplementary Files
Action
1. JATS XML
Download


Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.

Согласие на обработку персональных данных

 

Используя сайт https://journals.rcsi.science, я (далее – «Пользователь» или «Субъект персональных данных») даю согласие на обработку персональных данных на этом сайте (текст Согласия) и на обработку персональных данных с помощью сервиса «Яндекс.Метрика» (текст Согласия).