REAL-TIME TRACKING OF MOVING-OBJECTS WITH AN ACTIVE CAMERA

This article is concerned with the design and implementation of a syst em for real-time monocular tracking of a moving object using the two d egrees of freedom of a camera platform. Figure-ground segregation is b ased on motion without making any a priori assumptions about the objec t form. Using only the first spatiotemporal image derivatives, subtrac tion of the normal optical flow induced by camera motion yields the ob ject image motion. Closed-loop control is achieved by combining a stat ionary Kalman estimator with an optimal Linear Quadratic Regulator. Th e implementation on a pipeline architecture enables a servo rate of 25 Hz. We study the effects of time-recursive filtering and fixed-point arithmetic in image processing and we test the performance of the cont rol algorithm on controlled motion of objects. (C) 1998 Academic Press Limited.