Automatic drift elimination in probe microscope images based on techniques of counter-scanning and topography feature recognition

An experimentally proved method for the automatic correction of drift-distorted surface topography obtained with a scanning probe microscope (SPM) is suggested. Drift-produced distortions are described by linear transformations valid for the case of rather slow changing of the microscope drift velocity. One or two pairs of counter-scanned images (CSIs) of surface topography are used as initial data. To correct distortions, it is required to recognize the same surface feature within each CSI and to determine the feature lateral coordinates. Solving a system of linear equations, the linear transformation coefficients suitable for CSI correction in the lateral and the vertical planes are found. After matching the corrected CSIs, topography averaging is carried out in the overlap area. Recommendations are given that help both estimate the drift correction error and obtain the corrected images where the error does not exceed some preliminarily specified value. Two nonlinear correction approaches based on the linear one are suggested that provide a greater precision of drift elimination. Depending on the scale and the measurement conditions as well as the correction approach applied, the maximal error may be decreased from 8-25% to 0.6-3%, typical mean error within the area of corrected image is 0.07-1.5%. The method developed permits us to recover drift-distorted topography segments/apertures obtained by using feature-oriented scanning. The suggested method may be applied to any instrument of the SPM family.