Extrinsic Parameters Calibration of a Structured Light System Via Planar Homography Based on a Reference Solid

Structured light systems are widely used for three-dimensional shape reconstruction. The accuracy of the vision system parameters is of paramount importance for the accuracy of the 3D reconstruction. Often the extrinsic parameters change over time due to environmental conditions, in order to compensate for structured light parameters change, a self-recalibration method based on planar homography is proposed in this paper. The method uses the planar homography of corresponding points in two camera views to identify the relative position and orientation between the two cameras: one is the "real" camera, while the other is a reference solid considered as a "pseudo-camera". The laser plane of the structured light vision system is the constraining plane for the two cameras corresponding points. In order to assess the metrological quality of the method an evaluation of the accuracy of the 3D reconstruction is performed by a two step Monte Carlo simulation. In a first step it is estimated the extrinsic parameters accuracy obtained by self-recalibration. In a second step the intrinsic parameters accuracy and the extrinsic one previously computed are used to evaluate the accuracy of 3D shape reconstruction. In the final version of the paper, the overall accuracy will be verified experimentally by a proper calibration setup.