Adaptive Learning Image Tracking Algorithm Based on Characteristic Fusion

2021 ◽  
Vol 1993 (1) ◽  
pp. 012039
Author(s):  
Yu Feng ◽  
Chenlong Deng ◽  
Liyang Wu ◽  
Xingdi Wang ◽  
Jing Xu ◽  
...  
2017 ◽  
Vol 38 (4) ◽  
pp. 678-684 ◽  
Author(s):  
Chan-Yang Kuo ◽  
Ho-Chiao Chuang ◽  
Yi-Liang Zhou ◽  
Yu-Peng Wu ◽  
Jia-Chang Wang ◽  
...  

2012 ◽  
Vol 605-607 ◽  
pp. 1748-1752
Author(s):  
Dong Sen Si ◽  
Xiao Xu Wang ◽  
Chen Chen

Realize the goal quick and accurate location when the detection signal is weaker, puts forward a Fast Intelligent Tracking algorithm(FITA) with the Gain Adjustable. Wiht the goal image move in four quadrant detector, it’s size, location and output signal related to a linear probe interval. In this interval, a liner relation can be get between the spot position and the corresponding output signal. The target quick location algorithm is realized with this relation. On the other hand, the sum signal of the four quadrant detector is sampled in real-time, this sum is related to the sensor’s signal strength, then the change rate of the signal strength can be get, using adaptive learning ways to get the change trend of signal strength, and realized dynamic adjustment of the signal gain, in order to improve the signal resolution ratio when signal is weak, the tracking accuracy can be improved with this way. The experimental data show that FITA algorithm can rapidly locate position in the linear interval and accurately track through the dynamic change signal gain for the weaker signal. The track error can be improved to 1.9%.


2013 ◽  
Vol 722 ◽  
pp. 472-477
Author(s):  
Wu Xue Jiang ◽  
Zhi Ming Wang ◽  
Jian Feng Luo

Video surveillance system with image recognition and image tracking capabilities has a wide range of applications in the field of city protection, social security and public safety. In this paper, based on the analysis of image detection and tracking principles, image recognition and tracking algorithm based PID fuzzy control is proposed. First, the image on the basis of the gray-scale image is acquired by the video processor and the linearization processing, and then through the expansion operation of mathematical morphology the candidate region of the image recognition is taken, and then the final image region is obtained through the vertical positioning and horizontal positioning of the region Finally, through the PID control algorithm it is the image of the track. Simulation results show that the PID fuzzy control-based image recognition and tracking algorithm possesses good vehicle license plate image recognition and tracking performance and good practicability.


Sign in / Sign up

Export Citation Format

Share Document