Title :
Amphibious Target Tracking with Single PTZ Camera
Author :
Cheng-Chang Lien;Wang Jian Xiang
Author_Institution :
Dept. Comput. Sci. &
Abstract :
Recently, the researches about the video surveillance for constructing a safety environment and ecological investigation are highly demanded. In the video surveillance, the PTZ cameras are often applied to extract the close-up views by controlling the pan, tilt, and zoom operations in a large area. In this study, we develop a robust and amphibious target tracking with single PTZ camera in which the targets are tracked with the integration of object detection and SURF point tracking. First, we use the object detection method without background modeling to detect the interested object in the specified area. Second, the SURF-based target tracking scheme is used for object tracking. The proposed SURF-based target tracking scheme is established by integrating the SURF points updating and resampling methods to overcome the inaccurate SURF point matching problem under the dynamic scenes. Finally, the PTZ camera is controlled with the pan, tilt, and zoom operations to locate the target stably in the image center position. The experiment results show that the accuracy of SURF-based object tracking is about 86% and the efficiency can approach 15 fps.
Keywords :
"Target tracking","Object detection","Cameras","Feature extraction","Filtering","Reflection","Image color analysis"
Conference_Titel :
Intelligent Information Hiding and Multimedia Signal Processing (IIH-MSP), 2015 International Conference on
DOI :
10.1109/IIH-MSP.2015.73