会议专题

Real-Time Tracking for Multiple objects based on RGB Color Space

  Different algorithms based on image processing techniques have been used to detect and track the objects in video surveillance systems.In this paper,RGB colors based tracking for multiple objects are presented.For this,real-time tracking algorithm is proposed for the objects to navigate,and identify the current positions and their movements because each object has a color tagged on top.RGB color space is used to compute the threshold values of each color.Threshold values can be set depending on the object to be tracked.Linear model is applied for color discrimination.To remove unwanted objects from the frames,morphological processing is applied on binary image and then Blob analysis technique is used to facilitate the object detection.Our proposed algorithm is capable to track objects at full frame size (640×480 pixels) at frame rate (50fps).The high robustness and real time performance is confirmed by the experiments.

Real-time tracking Thresholding RGB colorspace Morphological operations Blob analysis

Kalisa Wilson

School of Information Science and Engineering,Xiamen University,Xiamen China

国际会议

2016IEEE第二届信息技术、网络、电子及自动化控制会议

重庆

英文

1051-1055

2016-03-20(万方平台首次上网日期,不代表论文的发表时间)