An object tracking method based on scenario-type hypothesis tracking in segmented multiple regions

In recent years, wireless communication technologies such as ZigBee have been developed. By constructing a wireless sensor network from infrared sensors with those wireless technologies, it is possible to track moving objects in a wide monitoring region. In this paper, we propose an object tracking method by dividing a wide monitoring region into small multiple regions called "micro cells". Pair of infrared sensors is placed at borders among micro cells to detect objects and their movement direction, and sensor information are gathered to a tracking server through wireless or wired networks. In our proposed method, the tracking server maintains virtual micro cells and virtual objects which indicate a variety of hypotheses of object trajectories. A virtual object is generated, moved, and removed in virtual micro cells appropriately according to sensor information and interactions among virtual objects, so that the tracking server keeps a number of object trajectory hypotheses for each object. When an object departs from the monitoring region, the virtual object with highest likelihood is selected as the estimated object trajectory. Through preliminary simulation evaluations, we confirm that tracking accuracy is improved up to 1.94 times higher by making virtual objects appropriately.