Control Systems and Computers, N2, 2020, Article 2
https://doi.org/10.15407/csc.2020.02.012
Control Systems and Computers, 2020, Issue 2 (286), pp. 12-22.
UDK 004.932.2
Kyyko Volodymyr M., PhD (Eng.), Senior Research Associate, International Research and Training Center for Information Technologies and Systems NAS and MES of Ukraine, Glushkov ave., 40, Kyiv, 03187, Ukraine, vkiiko@gmail.com,
Matsello Vyacheslav V., PhD (Eng.), Senior Research Associate, International Research and Training Center for Information Technologies and Systems NAS and MES of Ukraine, Glushkov ave., 40, Kyiv, 03187, Ukraine, matsello@gmail.com
Object Tracking at Video Monitoring
Introduction. Algorithms for visual object tracking are observed. In realistic scenarios each tracker is not superior in handling all interfering factors such as illumination and appearance variations, occlusions, changes of motion and scale of tracked object and so on. Therewith background subtraction algorithms are effective in handling motion and scale object changes, whereas appearance based ones alternatively in appearance, illumination and camera motion changes. Given the wide variety of aspects in tracking circumstances, development of trackers with co-operative actions of background subtraction and appearance based algorithms, specifically MOG (mixture of gaussians) and KCF (kernelized correlation filters) algorithms, is desirable.
Purpose of the article is to develop online tracking algorithm on the base of co-operative applying of the MOG and KCF algorithms.
Methods. Proposed tracker makes use of the KCF to find new position of tracked object in current frame and use of MOG to get subtractive image with subsequent correction of error object position on the base of integral representation of this image.
Results. Visual online tracking algorithm based on co-operative use of the MOG and KCF algorithms is developed. Testing results prove that the algorithm is more stable in comparison with the KCF in the case of abrupt changes of tracked object motion speed or direction. The algorithm is also more resistant to noise and illumination changes in comparison with the MOG algorithm.
Conclusions. The tracker handles appearance and scale changes of object if it does not move with other moving objects in the background and illumination variations are not large. When this occurs tracking is based mainly on the KCF algorithm. Further research for better handling this case is desirable.
Download full text! (On Ukrainian)
Keywords: Object tracking, tracking evaluation, tracking dataset, camera surveillance, image processing.
- Rout, R., 2013. A survey on object detection and tracking algorithms. 75 p.
- Yilmaz, A., Javed, O., Shah, M., 2006. “Object tracking: A survey”, ACM Computing Surveys (CSUR), 38 (4), pp. 1-45.
https://doi.org/10.1145/1177352.1177355 - Smeulders A., Chu D., Cucchiara R., Calderara S., Dehghan A., Shah M., 2014. “Visual tracking: an experimental survey”, IEEE Transactions on Pattern Analysis and Machine Intelligence, 36 (7), pp. 1442-1468.
https://doi.org/10.1109/TPAMI.2013.230 - Henriques, J. F., Caseiro, R., Martins, P., Batista, J., 2015. “High-speed tracking with kernelized correlation filters”, IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI), 37 (3), pp. 583-596.
https://doi.org/10.1109/TPAMI.2014.2345390 - Zivkovic, Z., van der Heijden, F., 2006. “Efficient adaptive density estimation per image pixel for the task of background subtraction”, Pattern recognition letters, 27 (7), pp. 773-780.
https://doi.org/10.1016/j.patrec.2005.11.005 - Grabner, H., Grabner, M., Bischof, H., 2006. “Real-time tracking via on-line boosting”, Proc. BMVC, 1, pp. 1-10.
https://doi.org/10.5244/C.20.6 - Babenko, B., Yang, M., Belongie, S., 2011. “Robust object tracking with online multiple instance learning”, IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 33 (8), pp. 1619-1632.
https://doi.org/10.1109/TPAMI.2010.226 - Viola, P., Jones, M., 2001. “Rapid object detection using a boosted cascade of simple features”. IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR). Kauai, USA, 1, pp. 511–518.
- Kalal, Z., Mikolajczyk, K., Matas, J., 2012. “Tracking-Learning-Detection”, IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 34 (7), pp. 1409-1422.
https://doi.org/10.1109/TPAMI.2011.239 - Bradski, G., 2000. “The OpenCV Library”. Dobb’s Journal of Software Tools, 25(11), pp. 122–125.
- Shlezinger, M. I., 1991. “Bystraya realizatsiya odnogo klassa lineynykh svertok”, Teoreticheskiye i prikladnyye voprosy raspoznavaniya izobrazheniy, Kyiv: IK AN USSR, pp. 61–69. (In Russian).
- Everingham, M., Gool, L.J.V., Williams, C.K.I., Winn, J. M., Zisserman, A., 2010. “The pascal visual object classes (VOC) challenge”, International Journal of Computer Vision (IJCV), 88 (2), pp. 303-338.
https://doi.org/10.1007/s11263-009-0275-4
Received 09.12.2019
