Fig. 1. The essential purpose of stitching is to determine the geometric relation for the subset data

Fig. 2. Photos of the hardware and the software interface in the NSLS-II stitching platform

Fig. 3. Photos of the X-ray flat mirror. To check the self-consistency of a measurement system, the SUT is usually tested in (a) A-to-B orientation and (b) B-to-Aorientation (SUT rotated 180° along its surface normal)

Fig. 4. Different sub-apertures (highlighted in a yellow rectangle) with (a) 600 pixel×150 pixel and (b) 256 pixel×64 pixel are used to measure this X-ray flat mirror in software stitching mode for comparison

Fig. 5. 30 repeated scans on the X-ray flat mirror show excellent stitching repeatability

Fig. 6. Measurement discrepancies of the central line (a) are at sub-nanometer level, and most of the discrepancies have an RMS value less than 0.15 nm (b)

Fig. 7. Similar results are obtained by different stitching interferometers at ESRF and NSLS-II using different sub-apertures in different mirror orientations on different dates

Fig. 8. x-slope values along the 2.5-mm-wide center line are calculated from the stitched data and compared with the NSLS-II NSP data. Some curves are vertically shifted for clarity

Fig. 9. Photo of the measured X-ray hyperbolic cylindrical mirror

Fig. 10. Stitched height maps of the X-ray hyperbolic cylindrical mirror surface (a) and the shape errors after removing the target shape (b)