Image processing is increasingly used for industrial measuring tasks. In this paper we will introduce a system which can estimate the pose of a camera as well as the 3D coordinates of particular object points. For input data the system uses several images containing landmarks taken from different camera poses. A nonlinear optimization routine is applied. Following local linearization a Kalman estimation in covariance form is performed. The result of the simulation and measurements prove the system´s efficiency.
Print ISSN: 0171-8096
Volume: 70, 10/2003
Pages: 475