Laser & Optoelectronics Progress, Volume. 56, Issue 3, 030401(2019)
Target Anti-Occlusion Algorithm Based on Correlation Filter
Fig. 1. Cyclic displacement of two dimensional images. (a) +30; (b) +15; (c) 0; (d) -15; (e) -30
Fig. 2. Occlusion judgment. (a) Be blocked when go to the right; (b) be blocked when go to the left; (c) response curve
Fig. 3. Occlusion judgment flow
Fig. 4. KCF algorithm flow
Fig. 5. Occlusion experiment of KCF. (a) Before occlusion; (b) when occlusion occurs; (c) after occlusion
Fig. 6. Anti-occlusion test. (a) Occlusion; (c) after the target appearing
Fig. 7. Forward and backward difference detection
Fig. 8. Feature detection
Fig. 9. Comparative experiment. (a) Walking; (b) motinas_Room105; (c) Suv; (d) David3; (e) Subway
Fig. 10. Threshold setting. (a) Delay; (c) loss
Fig. 11. Tracking result of Walking
Fig. 12. Tracking result of motinas_Room105
|
|
Get Citation
Copy Citation Text
Kaiyu Wang, Zhiguo Chen, Yi Fu. Target Anti-Occlusion Algorithm Based on Correlation Filter[J]. Laser & Optoelectronics Progress, 2019, 56(3): 030401
Category: Detectors
Received: Jul. 6, 2018
Accepted: Aug. 11, 2018
Published Online: Jul. 31, 2019
The Author Email: Chen Zhiguo (427533@qq.com)