Izvestiya Vysshikh Uchebnykh Zavedenii. Elektronika home

Publication of the journal

The section is currently being updated

Most existing video object tracking algorithms either implement short-term object tracking until the point of tracking failure without the possibility of its recovery, or detect the tracked object in each frame based on some pre-built object model, adapting to changes in its image. In the first case, a separate algorithm for object re-localization after a tracking failure is required; in the second, it is necessary to somehow compensate for tracking errors due to the problem of drift of the predicted bounding box when the object appearance changes. In this work, a method for solving the problem of robust object tracking is presented based on the universal approach to the building of adaptive tracking algorithms that include three components: tracking, learning and detection. The algorithm for object contour tracking has been developed to be used as a tracking component, as part of complex algorithm that implements the described approach. The use of this algorithm in the developed tracking system made it possible to increase the adaptability of tracking to severe changes in object appearance and to achieve robust object tracking in video sequences that are difficult to analyze. The functionality of the software developed on the basis of the proposed algorithm has been confirmed.

Key words: automatic tracking, object contour tracking, object detector, detector training, object detection
Published in: INFORMATION-COMMUNICATION TECHNOLOGIES
Bibliography link: Sotnikov А. V., Shipatov A. V., Savchenko Yu. V. An adaptive algorithm for object tracking in video images. Proc. Univ. Electronics, 2025, vol. 30, no. 1, pp. 94–106. https://doi.org/10.24151/1561-5405-2025-30-1-94-106
Financial source: The work has been supported by the Ministry of Education and Science of the Russian Federation (Agreement 075-10-2021-093, project RAI-RND-2126).

Alexander V. Sotnikov
National Research University of Electronic Technology (Russia, 124498, Moscow, Zelenograd, Shokin sq., 1); JSC “Zelenograd Innovation and Technology Center” (Russia, 124527, Moscow, Zelenograd, Solnechnaya Prwy, 8); Sirius University of Science and Techno

Andrey V. Shipatov
National Research University of Electronic Technology (Russia, 124498, Moscow, Zelenograd, Shokin sq., 1); JSC “Zelenograd Innovation and Technology Center” (Russia, 124527, Moscow, Zelenograd, Solnechnaya Prwy, 8); Sirius University of Science and Techno

1. Kalal Z., Mikolajczyk K., Matas J. Tracking-learning-detection // IEEE Trans. Pattern Anal. Mach.Intell. 2012. Vol. 34. No. 7. P. 1409-1422. DOI: 10.1109/TPAMI.2011.239

2. Lucas B. D., Kanade T. An iterative image registration technique with an application to stereo vision // Proceedings of the 7th International Joint Conference on Artificial Intelligence (IJCAI'81). San Francisco, CA: Morgan Kaufmann Publ., 1981. Vol. 2. P. 674-679.

3. Kalal Z., Matas J., Mikolajczyk K. P-N learning: Bootstrapping binary classifiers by structural constraints // 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. San Francisco, CA: IEEE, 2010. P. 49-56. DOI: 10.1109/CVPR.2010.5540231

4. Blake A., Isard M. Dynamic contour tracking // Active contours / A. Blake, M. Isard. London: Springer, 1998. P. 213-234. DOI: 10.1007/978-1-4471-1555-7_10

5. Cao S., Wang X. Visual contour tracking based on inner-contour model particle filter under complex background // J. Image Video Proc. 2019. Vol. 2019. Art No. 85. DOI: 10.1186/s13640-019-0487-7

6. A fast region-based active contour for non-rigid object tracking and its shape retrieval / H. Mewada, J. F. Al-Asad, A. Patel et al. // PeerJ Comput. Sci. 2021. Vol. 7. Art. ID: e373. DOI: 10.7717/peerj-cs.373 EDN: OAZQQM

7. Mansouri A.-R. Region tracking via level set PDEs without motion computation // IEEE Trans. Pattern Anal. Mach.Intell. 2002. Vol. 24. No. 7. P. 947-961. DOI: 10.1109/TPAMI.2002.1017621

8. Caselles V., Catté F., Coll T., Dibos F. A geometric model for active contours in image processing // Numer. Math. 1993. Vol. 66. P. 1-31. DOI: 10.1007/BF01385685

9. Osher S., Fedkiw R. Level set methods and dynamic implicit surfaces. New York: Springer, 2002. XIII, 273 p. DOI: 10.1007/b98879

10. Сотников А. В., Шипатов Е. А., Шипатов А. В. Алгоритм выделения объектов на изображениях на основе слияния суперпикселей // Изв. вузов. Электроника. 2023. Т. 28. № 1. С. 129-139. DOI: 10.24151/1561-5405-2023-28-1-129-139 EDN: VZAJVI
Sotnikov A. V., Shipatov E. A., Shipatov A. V. An algorithm for object extraction in images based on
superpixel fusion. Izv. vuzov. Elektronika = Proc. Univ. Electronics, 2023, vol. 28, no. 1, pp. 129–139. (In Russian). https://doi.org/10.24151/1561-5405-2023-28-1-129-139

11. TDL dataset // Yumpu [Электронный ресурс]. URL: https://www.yumpu.com/en/document/view/18189558/tld-dataset (дата обращения: 06.11.2024).

12. Разработка и создание универсальной открытой программно-аппаратной платформы для проектирования устройств обработки потокового видео для беспилотных летающих аппаратов мониторинга экологической ситуации и состояния природных объектов // ФЦП "Исследования и разработки по приоритетным направлениям развития научно-технологического комплекса России на 2014-2021 годы" [Электронный ресурс] / Министерство науки и высшего образования Российской Федерации. URL: https://fcpir.fcntp.ru/participation_in_program/contracts/05.605.21.0181/(дата обращения: 06.11.2024).
Development and establishment of a universal open hardware and software platform for designing streaming video processing devices for unmanned aerial vehicles monitoring ecological situation and natural objects condition. Ministry of Education and Science of the Russian Federation. FTsP “Issledovaniya i razrabotki po prioritetnym napravleniyam razvitiya nauchno-tekhnologicheskogo kompleksa Rossii na 2014–2021 gody”. Available at: https://fcpir.fcntp.ru/participation_in_program/contracts/05.605.21.0181/ (accessed: 06.11.2024).

Information about the publication

UDK: 004.932.2+519.688
Publication type: Scientific article
DOI: 10.24151/1561-5405-2025-30-1-94-106
RISC id: BKBAAG

JATS XML