Flexible correction of 3D non-linear drift in SPM measurements by data fusion

In this article a new offline method for correcting non-linear drift in all three dimensions (3D) is presented. Using this method, a sample region is measured in multiple sub-measurements, each with increased sampling distance and thus reduced measurement time. The datasets of the sub-measurements are aligned using a point-to-plane iterative closest point algorithm to reconstruct and correct the 3D drift. Afterwards, the corrected datasets are fused into a single dataset. Compared to conventional drift-correction methods, the new method has the advantages of drift correction in full 3D with higher temporal resolution, less extra measurement time and data redundancy, as well as high application flexibility (e.g. compatibility to non-raster sampling). However, the resulting dataset might have slightly decreased resolution. If a high-resolution low-drift dataset is required, the method can be applied in an extended way, where an additional high-resolution measurement is taken whose drift can be corrected by the aforementioned dataset generated by data fusion. Two experimental measurements and a simulation study have been carried out in a new low-noise 3D atomic force microscope, showing great application potential of the proposed method.