Evaluation of a novel algorithm to align and stitch adjacent measurements of long inner cylindrical surfaces with white light interferometry

This paper presents a novel algorithm based on least absolute method to align and stitch multiple cylindrical clouds of points measured using white light interferometry. The evaluation of the stitching algorithm was performed using several real measured clouds of points (COPs) and numerically simulated ones. Adjacent COP were processed to obtain the stitched one, and these results were compared with the original COP. In this way, the performance of the presented algorithm was evaluated and analyzed for several overlapped areas. It was possible to conclude that: (a) the better overlapped area between adjacent measurements was 80% of the total measured area when the metrological performance was taken into account; (b) the better overlapped area was 25% when the operational performance was analyzed; and (c) the standard deviation of the stitched COP were usually better than 100 nm. A brief description of a modified white light interferometer to measure in cylindrical coordinates is also presented.