Kalman filter incorporated model updating for real-time tracking

This paper describes a real-time tracking system which detects an object entering into the field of view of a camera and executes the tracking of the detected object by controlling a servo device so that a target object always lies at the center of the image frame. In order to detect and track a moving object, we basically apply a model matching strategy. We allow the model to vary dynamically during the tracking process so that it can assimilate the variations of shape and intensities of the target object. We also utilize a Kalman filter so that a tracking history can be encoded into the state parameters of the Kalman filter. The estimated state parameters of the Kalman filter is then used to reduce the search areas for model matching and to control the servo device