pp. 563-574
S&M2828 Research Paper of Special Issue https://doi.org/10.18494/SAM3616 Published: February 14, 2022 Camera Handoff for Multicamera Multiobject Tracking [PDF] Po-Tong Wang, Jia-Shing Sheu, and Jing-Han Lai (Received September 3, 2021; Accepted January 6, 2022) Keywords: camera handoff, background subtraction, human tracking
In recent years, the unmanned economy has begun to develop rapidly worldwide. Unmanned shops use technology to replace human personnel and increase efficiency. This technology has increased the demand for accurate and efficient camera surveillance systems. Camera handoff is a crucial step in the tasks of continuous target tracking and the maintenance of consistent cross-camera target marking in a multicamera surveillance system. In this study, we proposed a method of indoor multicamera handoff. To ensure continuous target tracking, the minimum number of required frames was maintained and at least one camera tracked each target. We employed the background subtraction approach to detect the target. Next, we used three trackability measures to evaluate the tracking object and trigger camera handoff accordingly, selecting the optimal camera for target tracking. The three measures considered were the resolution, occlusion, and distance to the edge of the camera’s field of view. When one of these reached a preset threshold with an increasing trend, the system triggered the camera handoff.
Corresponding author: Jia-Shing SheuThis work is licensed under a Creative Commons Attribution 4.0 International License. Cite this article Po-Tong Wang, Jia-Shing Sheu, and Jing-Han Lai, Camera Handoff for Multicamera Multiobject Tracking, Sens. Mater., Vol. 34, No. 2, 2022, p. 563-574. |