TY - GEN
T1 - Adaptive camera assignment and hand-off algorithm in multiple active camera surveillance system
AU - Lee, Chao Yang
AU - Lin, Shou Jen
AU - Lee, Chen Wei
AU - Yang, Chu-Sing
PY - 2010/11/15
Y1 - 2010/11/15
N2 - Recently, the pan-tilt-zoom (PTZ) camera is extensively used for surveillance in wide area environments. A current automatic system has been developed to continually track a person with the function of keeping the person centered in the field of view (FOV). Tracking moving objects has long been a key problem in computer vision. It is important in a wide variety of video surveillance applications and has been applied successfully in people's daily life, like human motion analysis, traffic monitoring, market guard, and in-house health care. When multiple cameras are used for tracking, the algorithm can automatically provide an optimal capacity and provide a solution to perform the camera assignment efficiently. This paper investigates the application of automatic methods for tracking individual across cameras via a surveillance network. Our approach is to track the individual and activate the function of surveillance in cameras to achieve seamless tracking with high quality image. We develop an emerging low computational complexity hand-off function that can automatically carry out the camera assignment and hand-off task. In the experiment, with a feasible solution for seamless tracking and real-time surveillance provided, our algorithm can perform the hand-off task efficiently and automatically in multiple active camera surveillance systems.
AB - Recently, the pan-tilt-zoom (PTZ) camera is extensively used for surveillance in wide area environments. A current automatic system has been developed to continually track a person with the function of keeping the person centered in the field of view (FOV). Tracking moving objects has long been a key problem in computer vision. It is important in a wide variety of video surveillance applications and has been applied successfully in people's daily life, like human motion analysis, traffic monitoring, market guard, and in-house health care. When multiple cameras are used for tracking, the algorithm can automatically provide an optimal capacity and provide a solution to perform the camera assignment efficiently. This paper investigates the application of automatic methods for tracking individual across cameras via a surveillance network. Our approach is to track the individual and activate the function of surveillance in cameras to achieve seamless tracking with high quality image. We develop an emerging low computational complexity hand-off function that can automatically carry out the camera assignment and hand-off task. In the experiment, with a feasible solution for seamless tracking and real-time surveillance provided, our algorithm can perform the hand-off task efficiently and automatically in multiple active camera surveillance systems.
UR - http://www.scopus.com/inward/record.url?scp=78149314660&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=78149314660&partnerID=8YFLogxK
U2 - 10.1109/ICMLC.2010.5580739
DO - 10.1109/ICMLC.2010.5580739
M3 - Conference contribution
AN - SCOPUS:78149314660
SN - 9781424465262
T3 - 2010 International Conference on Machine Learning and Cybernetics, ICMLC 2010
SP - 3015
EP - 3020
BT - 2010 International Conference on Machine Learning and Cybernetics, ICMLC 2010
T2 - 2010 International Conference on Machine Learning and Cybernetics, ICMLC 2010
Y2 - 11 July 2010 through 14 July 2010
ER -