International Journal of Computer Network and Information Security (IJCNIS)
IJCNIS Vol. 17, No. 2, 8 Apr. 2025
Cover page and Table of Contents: PDF (size: 983KB)
GMM-based Imbalanced Fractional Whale Particle Filter for Multiple Object Tracking in Surveillance Videos
PDF (983KB), PP.34-50
Views: 0 Downloads: 0
Author(s)
Avinash Ratre
1,*
Index Terms
Multiple Object Tracking, Particle Filter, Fractional Whale Optimization, Gaussian Mixture Model, Imbalanced Surveillance Video Data
Abstract
The imbalanced surveillance video dataset consists of majority and minority classes as normal and anomalous instances in the nonlinear and non-Gaussian framework. The normal and anomalous instances cause majority and minority samples or particles associated with high and low probable regions when considering the standard particle filter. The minority particles tend to be at high risk of being suppressed by the majority particles, as the proposal probability density function (pdf) encourages the highly probable regions of the input data space to remain a biased distribution. The standard particle filter-based tracker afflicts with sample degeneration and sample impoverishment due to the biased proposal pdf ignoring the minority particles. The difficulty in designing the correct proposal pdf prevents particle filter-based tracking in the imbalanced video data. The existing methods do not discuss the imbalanced nature of particle filter-based tracking. To alleviate this problem and tracking challenges, this paper proposes a novel fractional whale particle filter (FWPF) that fuses the fractional calculus-based whale optimization algorithm (FWOA) and the standard particle filter under weighted sum rule fusion. Integrating the FWPF with an iterative Gaussian mixture model (GMM) with unbiased sample variance and sample mean allows the proposal pdf to be adaptive to the imbalanced video data. The adaptive proposal pdf leads the FWPF to a minimum variance unbiased estimator for effectively detecting and tracking multiple objects in the imbalanced video data. The fractional calculus up to the first four terms makes the FWOA a local and global search operator with inherent memory property. The fractional calculus in the FWOA oversamples minority particles to be diversified with multiple imputations to eliminate data distortion with low bias and low variance. The proposed FWPF presents a novel imbalance evaluation metric, tracking distance correlation for the imbalanced tracking over UCSD surveillance video data and shows greater efficacy in mitigating the effects of the imbalanced nature of video data compared to other existing methods. The proposed method also outshines the existing methods regarding precision and accuracy in tracking multiple objects. The consistent tracking distance correlation near zero values provides efficient imbalance reduction through bias-variance correction compared to the existing methods.
Cite This Paper
Avinash Ratre, "GMM-based Imbalanced Fractional Whale Particle Filter for Multiple Object Tracking in Surveillance Videos", International Journal of Computer Network and Information Security(IJCNIS), Vol.17, No.2, pp.34-50, 2025. DOI:10.5815/ijcnis.2025.02.03
Reference
[1]M. S. Arulampalam, S. Maskell, N. Gordon, and T. Clapp, “A Tutorial on Particle Filters for Online Nonlinear/Non-Gaussian Bayesian Tracking,” IEEE Transactions on Signal Processing, vol. 50, no. 2, pp. 174–188, 2002, doi: 10.1109/78.978374.
[2]O. Cappe, S. J. Godsill, and E. Moulines, “An overview of existing methods and recent advances in sequential Monte Carlo,” Proceedings of the IEEE, vol. 95, no. 5, pp. 899–924, 2007, doi: 10.1109/JPROC.2007.893250.
[3]M. Isard and A. Blake, “CONDENSATION-Conditional Density Propagation for Visual Tracking,” Int J Comput Vis, vol. 29, no. 1, pp. 5–28, 1998.
[4]G. M. Rao and Ch. Satyanarayana, “Visual Object Target Tracking Using Particle Filter: A Survey,” International Journal of Image, Graphics and Signal Processing, vol. 5, no. 6, pp. 57–71, May 2013.
[5]F. Gustafsson et al., “Particle Filters for Positioning, Navigation, and Tracking,” IEEE Transactions on Signal Processing, vol. 50, no. 2, pp. 425–437, 2002.
[6]S. Wang, C. Shao, S. Xu, X. Yang, and H. Yu, “MSFSS: A whale optimization-based multiple sampling feature selection stacking ensemble algorithm for classifying imbalanced data,” AIMS Mathematics, vol. 9, no. 7, pp. 17504–17530, 2024, doi: 10.3934/math.2024851.
[7]E. M. Hassib, A. I. El-Desouky, L. M. Labib, and E. S. M. El-kenawy, “WOA + BRNN: An imbalanced big data classification framework using Whale optimization and deep neural network,” Soft comput, vol. 24, no. 8, pp. 5573–5592, Apr. 2020, doi: 10.1007/s00500-019-03901-y.
[8]S. Park, J. P. Hwang, E. Kim, and H. J. Kang, “A new evolutionary particle filter for the prevention of sample impoverishment,” IEEE Transactions on Evolutionary Computation, vol. 13, no. 4, pp. 801–809, 2009, doi: 10.1109/TEVC.2008.2011729.
[9]S. Mirjalili and A. Lewis, “The Whale Optimization Algorithm,” Advances in Engineering Software, vol. 95, pp. 51–67, May 2016, doi: 10.1016/j.advengsoft.2016.01.008.
[10]Y. Wu and T. S. Huang, “Robust Visual Tracking by Integrating Multiple Cues Based on Co-Inference Learning,” Int J Comput Vis, vol. 58, no. 1, pp. 55–71, 2004.
[11]J. Berclaz, F. Fleuret, E. Türetken, and P. Fua, “Multiple object tracking using k-shortest paths optimization,” IEEE Trans Pattern Anal Mach Intell, vol. 33, no. 9, pp. 1806–1819, 2011, doi: 10.1109/TPAMI.2011.21.
[12]X. Hai-Xia, W. Yao-Nan, Z. Wei, Z. Jiang, and Y. Xiao-Fang, “Multi-object visual tracking based on reversible jump Markov chain Monte Carlo,” IET Computer Vision, vol. 5, no. 5, pp. 282–290, Sep. 2011, doi: 10.1049/iet-cvi.2010.0086.
[13]M. Narayana, H. Nenavath, S. Chavan, and L. Koteswara Rao, “Intelligent visual object tracking with particle filter based on Modified Grey Wolf Optimizer,” Optik - International Journal for Light and Electron Optics, vol. 193, pp. 1–21, Sep. 2019, doi: 10.1016/j.ijleo.2019.06.013.
[14]M. Firouznia, J. A. Koupaei, K. Faez, G. A. Trunfio, and H. Amindavar, “Adaptive chaotic sampling particle filter to handle occlusion and fast motion in visual object tracking,” Digital Signal Processing: A Review Journal, vol. 134, pp. 1–13, Apr. 2023, doi: 10.1016/j.dsp.2023.103933.
[15]X. Yuqi, W. Yongjun, and Y. Fan, “A scale adaptive generative target tracking method based on modified particle filter,” Multimed Tools Appl, vol. 82, no. 20, pp. 31329–31349, Aug. 2023, doi: 10.1007/s11042-023-14901-4.
[16]S. Xu and H. Chen, “Visual Tracking via a Novel Adaptive Anti-occlusion Mean Shift Embedded Particle Filter,” Circuits Syst Signal Process, pp. 1–26, Feb. 2024, doi: 10.1007/s00034-024-02882-0.
[17]J. Li, Z. Cao, F. Liu, Y. Fu, X. Li, and F. Tian, “An adaptive biogeography-based optimization with integrated covariance matrix learning for robust visual object tracking,” Expert Syst Appl, vol. 234, pp. 1–24, Dec. 2023, doi: 10.1016/j.eswa.2023.121110.
[18]Y. Xiao and Y. Wu, “Robust visual tracking based on modified mayfly optimization algorithm,” Image Vis Comput, vol. 135, pp. 1–8, Jul. 2023, doi: 10.1016/j.imavis.2023.104691.
[19]M. Tian, Y. Bo, Z. Chen, P. Wu, and C. Yue, “Multi-target tracking method based on improved firefly algorithm optimized particle filter,” Neurocomputing, vol. 359, pp. 438–448, Sep. 2019, doi: 10.1016/j.neucom.2019.06.003.
[20]Z. Li, J. Chen, and J. Bi, “Multiple Object Tracking With Appearance Feature Prediction and Similarity Fusion,” IEEE Access, vol. 11, pp. 52492–52500, 2023, doi: 10.1109/ACCESS.2023.3279868.
[21]F. Wang, Y. Wang, J. He, F. Sun, X. Li, and J. Zhang, “Visual object tracking via iterative ant particle filtering,” IET Image Process, vol. 14, no. 8, pp. 1636–1644, Jun. 2020, doi: 10.1049/iet-ipr.2019.0967.
[22]S. Arora, T. Mathur, S. Agarwal, K. Tiwari, and P. Gupta, “Applications of fractional calculus in computer vision: A survey,” Neurocomputing, vol. 489, pp. 407–428, Jun. 2022, doi: 10.1016/j.neucom.2021.10.122.
[23]T. P. Morris, I. R. White, J. R. Carpenter, S. J. Stanworth, and P. Royston, “Combining fractional polynomial model building with multiple imputation,” Stat Med, vol. 34, no. 25, pp. 3298–3317, Nov. 2015, doi: 10.1002/sim.6553.
[24]M. S. Couceiro, R. P. Rocha, N. M. F. Ferreira, and J. A. T. Machado, “Introducing the fractional-order Darwinian PSO,” Signal Image Video Process, vol. 6, no. 3, pp. 343–350, Sep. 2012, doi: 10.1007/s11760-012-0316-2.
[25]E. J. Solteiro Pires, J. A. Tenreiro MacHado, P. B. De Moura Oliveira, J. Boaventura Cunha, and L. Mendes, “Particle swarm optimization with fractional-order velocity,” Nonlinear Dyn, vol. 61, no. 1–2, pp. 295–301, Jul. 2010, doi: 10.1007/s11071-009-9649-y.
[26]M. Dunnhofer, K. Simonato, and C. Micheloni, “Combining complementary trackers for enhanced long-term visual object tracking,” Image Vis Comput, vol. 122, pp. 1–10, Jun. 2022, doi: 10.1016/j.imavis.2022.104448.
[27]W. Dai, L. Jin, and G. Li, “Long-term adaptive tracking via complementary trackers,” IET Image Process, vol. 13, no. 9, pp. 1569–1577, Jul. 2019, doi: 10.1049/iet-ipr.2018.6142.
[28]Y. Xie, L. Peng, Z. Chen, B. Yang, H. Zhang, and H. Zhang, “Generative learning for imbalanced data using the Gaussian mixed model,” Applied Soft Computing Journal, vol. 79, pp. 439–451, Jun. 2019, doi: 10.1016/j.asoc.2019.03.056.
[29]M. Javvad ur Rehman, S. C. Dass, and V. S. Asirvadam, “A weighted likelihood criteria for learning importance densities in particle filtering,” EURASIP J Adv Signal Process, vol. 2018, no. 1, pp. 1–19, Dec. 2018, doi: 10.1186/s13634-018-0557-5.
[30]R. Gurajala, P. B. Choppala, J. S. Meka, and P. D. Teal, “A fast and unbiased minimalistic resampling approach for the particle filter,” in 2021 IEEE International Conference on Signal and Image Processing Applications (ICSIPA), Kuala Terengganu, Malaysia, Sep. 2021, pp. 227–232.
[31]C. Liu, Y. Wang, D. Zhou, and X. Shen, “Minimum-Variance Unbiased Unknown Input and State Estimation for Multi-Agent Systems by Distributed Cooperative Filters,” IEEE Access, vol. 6, pp. 18128–18141, Mar. 2018, doi: 10.1109/ACCESS.2018.2815662.
[32]“UCSD Anomaly Detection Dataset.” [Online]. Available: www.svcl.ucsd.edu/projects/anomaly/dataset.htm
[33]K. Bernardin and R. Stiefelhagen, “Evaluating multiple object tracking performance: The CLEAR MOT metrics,” EURASIP J Image Video Process, vol. 2008, pp. 1–10, 2008, doi: 10.1155/2008/246309.
[34]Y. Park, L. M. Dang, S. Lee, D. Han, and H. Moon, “Multiple object tracking in deep learning approaches: A survey,” Electronics (Switzerland), vol. 10, no. 19, pp. 1–31, Oct. 2021, doi: 10.3390/electronics10192406.
[35]S. Feng, J. Keung, P. Zhang, Y. Xiao, and M. Zhang, “The impact of the distance metric and measure on SMOTE-based techniques in software defect prediction,” Inf Softw Technol, vol. 142, pp. 1–14, Feb. 2022, doi: 10.1016/j.infsof.2021.106742.
[36]X. Yang and L. J. Latecki, “Weakly Supervised Shape Based Object Detection with Particle Filter,” in 2010 11th European Conference on Computer Vision (ECCV), Springer Berlin Heidelberg, 2010, pp. 757–770.