Uncertainty modeling in 3D SEM stereophotogrammetry

The scanning electron microscope (SEM) is widely used to acquire high resolution images. In order to reconstruct the third dimension of surface features, photogrammetry methods can be adopted. A specimen is imaged in the SEM acquiring two images, the stereo-pair, by scanning the same area from two different perspectives. The stereo-matching problem is solved by area- or feature-based methods implemented in commercial software. Piazzesi provided a first model for deriving surface topography from eucentric stereo-pairs. An uncertainty evaluation for the vertical elevation has been performed in a recent work for a cylindrical item. The aim of the present work is to extend the uncertainty evaluation to all surface coordinates. The proposed approach is based on the multivariate law of propagation of uncertainty (MLPU). Some preliminary results are also presented and discussed.