- 期刊
Improved MOSSE Tracking Algorithm Based On Dimensional Analysis
摘要
The classic MOSSE algorithm is often used for real-time target tracking. Aiming at the problem that the fixed filter update rate is difficult to adapt to the drastic change of the target appearance and the target is lost, an improved MOSSE tracking algorithm based on dimension analysis is proposed. This method uses Buckingham theorem to analyze the main factors that affect the update rate of the filter, establishes an update rate prediction model and automatically reloads the update rate. The improved algorithm can improve the robustness and tracking quality of target tracking.
參考文獻
K.Fukunaga, L.D. Hostetler. The estimation of the gradient of a density function, with application in pattern recognition.IEEE Transactions on Information Theory, 1975, 21: 32-40.
Gordon N, Salmond D. Novel approach to non-linear and non-Gaussian Bayesian state estimation[J]Proceedings of Institute Electric Engineering, 1993, 140(2):107-113
Kalal, Z., Mikolajczyk, K. and Matas, J. (2012) Track-ing-Learning-Detection. IEEE Transactions on Pattern Analysis and Machine Intelligence, 34, 1409-1422.
Bolme D S, Beveridge J R, Draper B A, et al. Visual object tracking using adaptive correlation filters [A]Computer Vision and Pattern Recognition[C].2010:2544-2550.
Piotr Kokoszka,Matthew Reimherr,Nikolas Wölfing. A randomness test for functional panels[J].Journal of Multivariate Analysis,2016,151.