D.A. Akimov1, E.N. Matyuhina2, A.W. Volosova3, S.W. Panjukova4, P.A. Berezin5, E.O. Gurianova6

1, 2, 5, 6 RTU MIREA (Moscow, Russia)
3, 4 Bauman Moscow State Technical University (Moscow, Russia)
1 akimov_d@mirea.ru, 2 matyuhina@mirea.ru, 3 volosova@.bmstu.ru,
4 Panyukova@bmstu.ru, 5 pav200293@gmail.ru, 6 guryanova@mirea.ru

Formulation of the problem: Many methods are used to solve the problem of landing on a runway. For solution of practical tasks, such as detection of lines on airfield images, key methods such as Gaussian blur, Canny detector and Hough transform, as well as their integration are used. However, the results deteriorate in low contrast or atmospheric interference, which requires additional filtering or adaptive algorithms.

Purpose: Adaptation of classical computer vision algorithms for automatic landing tasks, as well as the proposal of hybrid approaches that combine these methods with neural network models to automate parameter tuning.

Results: The paper proposes an integration of image processing methods, such as Gaussian blur, Canny detector, and Hough transform, to detect key landmarks (e.g., runway lines) in aviation synthetic vision systems. Through a Python implementation using OpenCV, the paper demonstrates the ability to accurately extract lines from images, including partial occlusion and noise conditions. A framework has been formed for further optimization of algorithms, including the development of adaptive filters and the integration of machine learning for stable operation in non-ideal conditions.

Practical significance: The obtained results make it possible to reduce the risk of accidents and improve flight safety by increasing the accuracy of object boundary detection, which is critical for navigation systems and airfield infrastructure monitoring.  

Akimov D.A., Matyuhina E.N., Volosova A.W., Panjukova S.W., Berezin P.A., Gurianova E.O. A multi-level information processing system in the UAV landing control unit using simulation modeling. Highly Available Systems. 2025. V. 21. № 4. P. 71−85. DOI: https://doi.org/10.18127/j20729472-202504-07 (in Russian)

1. Shchekacheva A.V., Kulyas O.L. Vydeleniye liniy na izobrazheniyakh s pomoshch'yu preobrazovaniy Khafa. Alleya Nauki. 2019. № 1. S. 34–46.
2. Forsayt D., Pons D. Komp'yuternoye zreniye. Sovremennyy podkhod. M.: Vil'yams. 2004. 928 s.
3. Soyfer V.A. Metody komp'yuternoy obrabotki izobrazheniy. M.: Fizmatlit. 2003. 784 s.
4. Drapey Ya.R., Malyshev V.I., Meleshko V.K. Sovremennyye metody i algoritmy dlya vosstanovleniya i uluchsheniya kachestva razmytykh izobrazheniy. Molodoy issledovatel' Dona. 2024. № 4. S. 22–29.
5. Pakhomova O.A., Kravets O.YA. Sravnitel'nyy analiz gradiyentnykh metodov vydeleniya kontura ob"yekta na izobrazhenii. Vestnik Voronezhskogo gosudarstvennogo tekhnicheskogo universiteta. 2018. № 1. S. 37–42.
6. Martyshkin A.I., Bershadskaya Ye.G., Markin Ye.I., Zuparova V.V. Modifikatsiya metoda vydeleniya konturov ob"yekta v intellektual'nykh sistemakh. Trudy ISP RAN. 2022. № 6. S. 127–136.
7. Semenov A.N. Obnaruzheniye radiolokatsionnykh tseley s pomoshch'yu preobrazovaniya Khafa. Nauka i Obrazovaniye. MGTU im. Baumana. 2014. № 12. S. 619–632.
8. Cabrera-Ponce A.A., Martinez-Carranza J. Hough Transform-Based Runway Detection for UAV Autonomous Landing. IEEE Transactions on Aerospace and Electronic Systems. 2021. V. 57. № 3. P. 1465–1478. DOI 10.1109/TAES.2021.3051965
9. Yang L., Qi J., Song D., Xiao J. Vision-Based UAV Landing Using Line Detection and Tracking. Sensors. 2020. V. 20. № 12. P. 3556. DOI 10.3390/s20123556
10. Garcia-Pardo P.J., Sukhatme G.S., Montgomery J.F. Towards Vision-Based Safe Landing for UAVs in Unstructured Environments. Journal of Intelligent & Robotic Systems. 2019. V. 94. № 1. P. 367–385. DOI 10.1007/s10846-018-0891-8
11. Kong W., Zhang D., Wang X., Chen J. Autonomous Landing of UAVs on Moving Platforms via Hough Transform and Kalman Filter. IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Madrid, 2018. P. 1–9. DOI 10.1109/IROS.2018.8593889
12. Lyu Y., Bai L., Huang X. A Robust Runway Detection Algorithm Based on Hough Transform and Region Growing for UAV Landing. International Journal of Advanced Robotic Systems. 2017. V. 14. № 5. DOI 10.1177/1729881417731088
13. Meng X., Wang W., Han J. Vision-Based UAV Autonomous Landing with Improved Hough Transform. Chinese Journal of Aeronautics. 2016. V. 29. № 4. P. 976–985. DOI 10.1016/j.cja.2016.06.005
14. Gorbunov M.I., Lebedev S.V. Analiz tochnosti detektsii VPP s ispol'zovaniyem preobrazovaniya Khafa. Trudy MAI. 2019. № 106. S. 12–23.
15. Klimov D.V., Fedorov R.N. Avtomaticheskaya posadka BPLA na osnove vizual'noy navigatsii. Avtomatika i telemekhanika. 2018. № 5. S. 78–92.
16. Belov A.A., Shishkin V.V. Sintez algoritmov upravleniya posadkoy BPLA s ispol'zovaniyem tekhnicheskogo zreniya. Nauchnyy vestnik MGTU GA. 2017. № 256. S. 45–58.
17. Volosova A.V., Matyukhina Ye.N., Morozov Ye.A. Realizatsiya bezopasnogo upravleniya svetoforami s ispol'zovaniyem neyromorfnoy vychislitel'noy bazy na osnove nechetkikh grafov. Computational Nanotechnology. 2025. T. 12. № 1. S. 11–16. DOI 10.33693/2313-223X-2025-12-1-11-16