Quality of 3D digitised points obtained with non-contact optical sensors

Today digitising systems deliver large sets of data points which are representative of an object surface, in a relatively short time. These digitised data are generally exploited by applications such as reverse engineering, inspection or free form copying. However, the quality of the point exploitation is widely linked to the point quality, i.e. to the quality of the digitising process. To answer a part of this problem, we propose a method to locally evaluate the uncertainty of location of a 3D point in the whole measuring space. This uncertainty integrates errors due to digitising noise and optical phenomena. In particular, the use of noisy data in the calibration method implies a dispersion on the resulting 3D point as obtained from its 2D CCD location.