Access the full text.
Sign up today, get DeepDyve free for 14 days.
Loading next page...
/lp/de-gruyter/multi-scale-target-tracking-algorithm-with-kalman-filter-in-ANsP3LCCum
References
References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.
- Publisher
- de Gruyter
- Copyright
- © 2017 Yichen Duan et al., published by Sciendo
- eISSN
- 2470-8038
- DOI
- 10.21307/ijanmc-2017-030
- Publisher site
- See Article on Publisher Site
Abstract
AbstractReal-time Compressive Tracking (CT) uses the compression sensing theory to provide a new research direction for the target tracking field. The algorithm is simple, efficient and real-time. But there are still shortcomings: tracking results prone to drift phenomenon, cannot adapt to tracking the target scale changes. In order to solve these problems, this paper proposes to use the Kalman filter to generate the distance weights, and then use the weighted Bayesian classifier to correct the tracking position, and perform multi-scale template acquisition in the determined position to adapt to the changes of the target scale. Finally, introducing the adaptive learning rate while updating to improve the tracking effect.. Experiments show that the improved algorithm has better robustness than the original algorithm on the basis of maintaining the original algorithm real-time.
Journal
International Journal of Advanced Network Monitoring and Controls – de Gruyter
Published: Jan 1, 2017
Keywords: compression sensing; CT; multi-scale; Kalman filter