Servo tracking of muti-color moving target based on target reference histogram update
This paper presents a novel enhanced Camshift algorithm for muti-color moving target tracking. Target reference histogram update is combined with Camshift to create a robust real-time tracking system . After the background is reconstructed and object is extracted in HSV colorspace , the HSV histogram of tracked target is created , called target reference histogram(TRH). During the tracking ,the TRH is updated through learning the value(intensity) of within and out of the region of interest(ROI), to minimizing the impact of environment light changing. A closed loop control model based on speed regulation is applied to drive the active camera to keep the target in the center of the camera angle.The experiment result shows that our system can track the muti-color moving target well in the complex dynamic environment by using the enhanced algorithm, even when the environment light changes a lot. The system is computationally efficient and can run in real-time completely.
Camshift Object histogram Histogram update Servo control Real-time tracking
Songxiao Cao Xuanyin Wang
The State Key Laboratory of Fluid Power Transmission and Control, Zhejiang University, Hangzhou, China
国际会议
杭州
英文
481-485
2009-04-08（万方平台首次上网日期，不代表论文的发表时间）