We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Object Tracking via Background-Aware Correlation Filter with Elliptical Search Area.
- Authors
Wencheng Yu
- Abstract
The utilization of background-aware correlation filtering (BACF) has been a topic of significant research interest due to its ability to enhance both the accuracy and speed of target tracking algorithms. Despite its advantages, the BACF algorithm struggles to overcome tracking drift in complex scenarios, such as those involving fast motion and occlusion, since it does not take into account the motion state of the target. To address this limitation, we propose a novel algorithm called redesigning the search area of background-aware correlation filtering (RSABACF). This algorithm considers the complete motion state of the target and uses the Kalman filter to formulate an elliptical search area. Furthermore, the proposed algorithm can dynamically adjust the search range and angle based on the target’s motion state to improve the background information in the motion direction and reduce it in the non-motion direction. As a result, the algorithm exhibits an improved ability to differentiate between target data and background data. Our proposed algorithm has been evaluated on the publicly available dataset OTB2015 and compared with mainstream tracking algorithms. The results show that our algorithm outperforms the BACF algorithm by 3.7% and 2.3% in terms of precision and success rates, respectively. Consequently, the proposed algorithm demonstrates its efficacy in handling complex tracking scenarios and exhibits high levels of robustness.
- Subjects
TRACKING algorithms; TRACKING radar; KALMAN filtering; ALGORITHMS
- Publication
Engineering Letters, 2023, Vol 31, Issue 4, p1747
- ISSN
1816-093X
- Publication type
Article