AN ADAPTIVE KALMAN FILTER FOR VIDEO-BASED TRACKING WITH DYNAMIC ADAPTATION OF COVARIANCE MATRICES
DOI:
https://doi.org/10.68302/std2026.vol2.204Keywords:
Adaptive Kalman filter, video-based object tracking, covariance matrix estimation, innovation-based adaptation, state estimation, measurement noiseAbstract
This paper presents an adaptive Kalman filter for video-based object tracking, in which the process and measurement covariance matrices are updated dynamically in real time. The proposed approach is based on a linear stochastic model without a control input, which is justified by the characteristics of video-based observation. The state vector includes position, velocity, and acceleration components, enabling a more accurate representation of motion dynamics. The main contribution of the study is the development of an adaptive algorithmic structure for estimating the covariance matrices based on innovation statistics. This approach improves consistency between the model and measurements under varying conditions and in the presence of noise. The effectiveness of the proposed method is evaluated through numerical simulation. The results demonstrate a significant reduction in the root mean square error (RMSE) in estimating position and velocity compared to unfiltered measurements and the classical Kalman filter. These results confirm the applicability of the proposed approach to video-based measurement systems.
Supporting Agencies
The authors acknowledge the support of the European Regional Development Fund, within the Operational Programme “Research, Innovation and Digitalization Programme for Intelligent Transformation 2021–2027”, through Project No. BG16RFPR002-1.014-0006 for the experimental investigations and measurement equipment, and through Project No. BG16RFPR002-1.014-0005 for the theoretical development and analytical framework of the study.Downloads
References
[1] J. R. SM and G. Augasta, “Review of recent advances in visual tracking techniques,” Multimedia Tools Appl., vol. 80, no. 16, pp. 24185–24203, 2021.
[2] M. A. Awal, M. A. R. Refat, F. Naznin, and M. Z. Islam, “A particle filter based visual object tracking: A systematic review of current trends and research challenges,” Int. J. Adv. Comput. Sci. Appl., vol. 14, no. 11, 2023.
[3] V. Zimparov, V. Petkov, and H. Hristov, “Optimal spacings for channels with Hagen–Poiseuille fluid-flow and mass transfer,” Thermal Science, vol. 25, spec. issue 2, pp. 295–301, 2021.
[4] I. Malakov, V. Zaharinov, S. Nikolov, and R. Dimitrova, “Computer-aided choosing of an optimal structural variant of a robot for extracting castings from die casting machines,” Actuators, vol. 12, p. 363, 2023.
[5] P. Mitev, “Development of a system for flexible feeding of parts with robot and machine vision,” Eng. Proc., vol. 104, p. 84, 2025.
[6] E. Dilek and M. Dener, “Computer vision applications in intelligent transportation systems: A survey,” Sensors, vol. 23, no. 6, p. 2938, 2023.
[7] Y. K. Lin, C. Y. Tu, L. Kurosawa, J. H. Liu, Y. Z. Wang and D. Roy, “Applications of computer vision in transportation systems: A systematic literature review,” in SHS Web Conf., vol. 194, p. 01004, 2024.
[8] N. Madzharov and N. Hinov, “Analysis and design of resonant DC/AC converters with energy dosing for induction heating,” Energies, vol. 16, no. 3, p. 1462, 2023.
[9] J. D. Trivedi, S. D. Mandalapu, and D. H. Dave, “Vision-based real-time vehicle detection and vehicle speed measurement using morphology and binary logical operation,” J. Ind. Inf. Integr., vol. 27, p. 100280, 2022.
[10] D. McDuff, J. Hernandez, E. Wood, X. Liu, and T. Baltrusaitis, “Advancing non-contact vital sign measurement using synthetic avatars,” arXiv preprint, arXiv:2010.12949, 2020.
[11] P. Kadam, G. Fang, and J. J. Zou, “Object tracking using computer vision: A review,” Computers, vol. 13, no. 6, p. 136, 2024.
[12] G. Iliev, H. Hristov, and D. Dimitrov, “Experimental study of the frequency characteristics of an electropneumatic tracking system with high speed pneumatic valves and PWM control,” in Proc. 16th Int. Sci. Conf. Environment Technology Resources, vol. 4, pp. 111–115, 2025.
[13] Z. Guan et al., “Multi-object tracking review: Retrospective and emerging trend,” Artif. Intell. Rev., vol. 58, no. 8, p. 235, 2025.
[14] G. Dinev, I. Malakov, and D. Dotsev, “CAD optimal design, documentation and automated assembly of mechanical product,” Adv. Mater. Res., vols. 463–464, pp. 1202–1205, 2012.
[15] W. Zhang et al., “Facing challenges: A survey of object tracking,” Digit. Signal Process., vol. 161, p. 105082, 2025.
[16] P. Mitev, “Development of a system for the active orientation of small screws,” Eng. Proc., vol. 70, p. 55, 2024.
[17] A. Sharma, L. Wang, and Y. Gao, “Rethinking object detection and tracking,” in Proc. Web Conf. Workshops, 2026.
[18] G. Iliev and H. Hristov, “Mathematical models of the static flow characteristic curve of electropneumatic proportional control valves,” J. Phys. Conf. Ser., vol. 3145, p. 012024, 2025.
[19] S. Akhlaghi, N. Zhou, and Z. Huang, “Adaptive adjustment of noise covariance in Kalman filter for dynamic state estimation,” in IEEE Power Energy Soc. Gen. Meeting, pp. 1–5, 2017.
[20] I. Malakov, Tz. Georgiev, V. Zaharinov, A. Tzokev and V. Tzenov, “Demand modeling for the optimization of size ranges,” in Proc. 26th DAAAM Int. Symp., pp. 435–444, 2015.
[21] P. Mitev, “Development of a training station for the orientation of dice parts with machine vision,” Eng. Proc., vol. 70, p. 57, 2024.
[22] G. Iliev and H. Hristov, “Selection and verification of an accurate mathematical model of proportional directional control valves,” J. Phys. Conf. Ser., vol. 3145, p. 012025, 2025.
[23] Y. Huang et al., “A novel adaptive Kalman filter with inaccurate process and measurement noise covariance matrices,” IEEE Trans. Autom. Control, vol. 63, no. 2, pp. 594–601, 2017.
[24] P. Mitev and I. Malakov, “Development of a system for automatic feeding and orientation of cylindrical parts,” AIP Conf. Proc., vol. 3063, p. 060017, 2024.
[25] X. R. Li and V. P. Jilkov, “Survey of maneuvering target tracking. Part V: Multiple-model methods,” IEEE Trans. Aerosp. Electron. Syst., vol. 41, no. 4, pp. 1255–1321, 2005.
[26] G. Iliev and H. Hristov, “Modelling and simulation of dynamic processes in an electropneumatic positioning system with high-speed valves and PWM control,” J. Phys. Conf. Ser., vol. 3127, p. 012012, 2025.
[27] X. Yu and Z. Meng, “Robust Kalman filters with unknown covariance of multiplicative noise,” IEEE Trans. Autom. Control, vol. 69, no. 2, pp. 1171–1178, 2023.
[28] D. Dichev, H. Koev, T. Bakalova, and P. Louda, “An algorithm for improving the accuracy of systems measuring parameters of moving objects,” Metrol. Meas. Syst., vol. 23, no. 4, pp. 555–565, 2016.
[29] Y. Cheng et al., “Self-tuning process noise in variational Bayesian adaptive Kalman filter for target tracking,” Electronics, vol. 12, no. 18, p. 3887, 2023.
[30] D. Dichev, D. Diakov, and R. Dicheva, “Method for increasing the accuracy of linear measurements based on a measurement-computational approach,” in AIP Conf. Proc., vol. 2505, 2022.
[31] J. O. D. A. Limaverde Filho, E. L. Fortaleza, J. G. Silva, and M. C. M. M. De Campos, “Adaptive Kalman filtering for closed-loop systems based on the observation vector covariance,” Int. J. Control, vol. 95, no. 7, pp. 1731–1746, 2022.
[32] L. Lazov, E. Teirumnieks, I. Draganov, and N. Angelov, “Numerical modeling and simulation for laser beam welding of ultrafine-grained aluminium,” Laser Physics, vol. 31, no. 6, Art. no. 066001, 2021.
[33] D. Pulov, P. Tsvyatkov, “Optical Systems for Reducing the Divergence of Laser Beams”, in. Proceedings of the 14th International Scientific and Practical Conference Environment. Technology. Resources. Rezekne, Latvia, Volume 3, 339-343, 2023, DOI: 10.17770/etr2023vol3.7217.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Dimitar Dichev, Iliya Zhelezarov, Tsanko Karadzhov, Kaloyan Libchev
This work is licensed under a Creative Commons Attribution 4.0 International License.
