Optimization of Parametric Object Video Stream Tracking Algorithms
https://doi.org/10.1134/S2304487X19030052
Abstract
An increase in the efficiency of algorithms of object tracking in video streams through the parametric optimization of these algorithms has been discussed. In contrast to most of such studies, accuracy and performance parameters are taken into account in one objective optimization function. There are five different results of object tracking in video streams: correct tracking, false object tracking, premature object loss, finding an object with a delay, and double object detection after premature loss. It has been shown that the parametric optimization makes it possible to increase the accuracy and performance of template matching tracking by factors of 4.7 and 2.2, respectively. At the same time, it has been found that the absence of parametric optimization can lead in practice to the refusal of efficient modifications. For example, template adaptation modification increases the template tracking quality only in 10% of the cases and decreases it in other 90 % of the cases. Particular parameters of the algorithm for template object tracking in video streams have been recommended for computer vision systems.
Keywords
parametric optimization,
object tracking in video stream,
template matching,
computer vision,
video stream
Supporting agencies: The study was carried out within the framework of project No. 25.2911.2017/4.6 “Approbation of technology for reducing the risk of occurrence and reducing the consequences of man-made disasters by minimizing the influence of the human factor on the reliability and trouble-free operation of nuclear power plants and other hazardous facilities” in accordance with the competitive part of the state assignment of the NRU MEPhI for 2017-2019.
About the Author
A. D. Egorov
National Research Nuclear University MEPhI (Moscow Engineering Physics Institute)
Russian Federation
Russian Federation
115409
Moscow
References
1. Bradski G., Kaehler A. Learning OpenCV: Computer vision with the OpenCV library. “O’Reilly Media, Inc.”, 2008.
2. Al-Kaff A. et al. Survey of computer vision algorithms and applications for unmanned aerial vehicles // Expert Systems with Applications. 2018. Vol. 92. P. 447–463.
3. Buch N., Velastin S. A., Orwell J. A review of computer vision techniques for the analysis of urban traffic // IEEE Transactions on Intelligent Transportation Systems. 2011. Vol. 12. № 3. P. 920–939.
4. Alyushin M. V., Alyushin V. M., Kolobashkina L. V. “Methodological aspects of automated forecasting of emergency situations of technogenic origin”. Questions of psychology. 2016. V. 2. P. 83–90.
5. Alyushin M. V., Alyushin A. V., Belopolskiy V. M., Kolobashkina L. V., Ushakov V. D. “Optical Technologies for Monitoring Systems of the Current Functional State of the Operational Management Personnel of Nuclear Power Facilities”. Global Nuclear Safety. 2013. V. 2. № 7. P. 69–77.
6. Joskowicz L. Future Perspectives on Statistical Shape Models in Computer-Aided Orthopedic Surgery: Beyond Statistical Shape Models and on to Big Data // Computer Assisted Orthopaedic Surgery for Hip and Knee. Springer, Singapore, 2018. P. 199–206.
7. Kutsiy N. N. Parametricheskaya optimizaciya sistem s integral’noj shirotno-impul’snoj modulyaciej s ispol’zovaniem bespoiskovogo gradientnogo algoritma [Parametric optimization of systems with integral pulse-width modulation using a searchless gradient algorithm]. Instruments and Systems: Monitoring, Control, and Diagnostics, 2012, vol. 1, pp. 2–6.
8. Kohno A. G. Mnogokriterial’naya parametricheskaya optimizaciya sudovyh avtomatizirovannyh sistem. Kand. Diss. [Multi-criteria parametric optimization of ship automated systems. Kand. Diss.]. Saint-Petersburg, 2012.
9. Sklyarko A. A. Metody resheniya zadachi popiksel’noj s-approksimacii mul’titonovyh izobrazhenij i ih optimizaciya. Kand. Diss. [methods for solving the pixix s-approximation multiton images and their optimization. Kand. Diss.], Rostov-on-Don, 2012.
10. Dunai Dunai L. et al. Euro banknote recognition system for blind people // Sensors. 2017. Vol. 17. № 1. P. 184.
11. Bruce B. R., Aitken J. M., Petke J. Deep parameter optimization for face detection using the viola-jones algorithm in OpenCV // International Symposium on Search Based Software Engineering. Springer, Cham. 2016. P. 238–243.
12. Egorov A. D., Shtanko A. N., Minin P. E. Selection of Viola–Jones algorithm parameters for specific conditions // Bulletin of the Lebedev Physics Institute. 2015. Vol. 42. № 8. P. 244–248.
13. Smeulders A. W. M. et al. Visual tracking: An experimental survey // IEEE transaction on pattern analysis and machine intelligence. 2014. Vol. 36. № 7. P. 1442–1468.
14. Oron S. et al. Locally orderless tracking // International Journal of Computer Vision. 2015. Vol. 111. № 2. P. 213–228.
15. Briechle K., Hanebeck U. D. Template matching using fast normalized cross correlation // Optical Pattern Recognition XII. International Society for Optics and/Photonics, 2001. Vol. 4387. P. 95–103.
16. Ross D.A. et al. Incremental learning for robust visual tracking // International journal of computer vision. 2008. Vol. 77. № 1–3. P. 125–141.
17. Isard M., Blake A. A mixed-state condensation tracker with automatic model-switching // Computer Vision, 1998. Sixth International Conference on. IEEE, 1998. P. 107–112.
18. Mei X., Ling H. Robust visual tracking using ℓ 1 minimization // Computer Vision, 2009 IEEE 12th International Conference on. IEEE, 2009. P. 1436–1443.
19. Ayvaci A., Raptis M., Soatto S. Sparse occlusion detection with optical flow // International journal of computer vision. 2012. Vol. 97. № 3. P. 322–338.
20. Nguyen H. T., Smeulders A. W. M. Robust tracking using foreground-background texture discrimination // International Journal of Computer Vision. 2006. Vol. 69. № 3. P. 277–293.
21. Godec M., Roth P. M., Bischof H. Hough-based tracking of non-rigid objects // Computer Vision and Image Understanding. 2013. Vol. 117. № 10. P. 1245–1256.
22. Salti S., Cavallaro A., Di Stefano L. Adaptive appearance modeling for video tracking: Survey and evaluation // IEEE Transactions on Image Processing. 2012. Vol. 21. № 10. P. 4334–4348.
23. Chan T. F., Vese L. A. Active contours without edges // IEEE Transactions on image processing. 2001. Vol. 10. № 2. P. 266–277.
24. Briechle K., Hanebeck U. D. Template matching using fast normalized cross correlation // Optical Pattern Recognition XII. International Society for Optics and Photonics, 2001. Vol. 4387. P. 95–103.
25. Shen X. et al. Detecting and aligning faces by image retrieval // Computer Vision and Pattern Recognition (CVPR), 2013 IEEE Conference on. IEEE, 2013. P. 3460–3467.
26. Leibe B., Leonardis A., Schiele B. Robust object detection with interleaved categorization and segmentation // International journal of computer vision. 2008. Vol. 77. № 1–3. P. 259–289.
27. Ballard D. H. Generalizing the Hough transform to detect arbitrary shapes // Readings in computer vision. 1987. P. 714–725.
28. Felzenszwalb P. F. et al. Object detection with discriminatively trained part-based models // IEEE transactions on pattern analysis and machine intelligence. 2010. Vol. 32. № 9. P. 1627–1645.
29. Chen D. et al. Joint cascade face detection and alignment // European Conference on Computer Vision. Springer, Cham, 2014. P. 109–122.
30. Martins P., Caseiro R., Batista J. Generative face alignment through 2.5 d active appearance models // Computer Vision and Image Understanding. 2013. Vol. 117. № 3. P. 250–268.
31. Samoylov A. S. et al. Face tracking for a system of collecting statistics on visitors and quality assessment of its functioning // Journal of Theoretical & Applied Information Technology. 2015. Vol. 71. № 3.
Review
For citations:
Egorov A.D. Optimization of Parametric Object Video Stream Tracking Algorithms. Vestnik natsional'nogo issledovatel'skogo yadernogo universiteta "MIFI". 2019;8(4):350-360. (In Russ.) https://doi.org/10.1134/S2304487X19030052
Views:
100
Email this article 