Practical Electron Microscopy and Database

Using energy-filtering-based techniques, e.g. elemental mapping and thickness measurement by EFTEM, it is possible to do complex data analysis. However, whenever several images are required to extract certain information, sample drift between the series of exposures are induced. The spatial drift mainly affects the quality of the final map in two ways:
      i) Each image in the image series can be taken from slightly different regions of the sample. This effect is often too big to be ignored.
      ii) The information within an individual image is blurred. This effect can often be negligible if image acquisition times and sample drift are small.

In order to obtain artifact-free information, this spatial drift has to be eliminated by shifting the images to find the best overlap. This process can be done by software or manual alignments. In software-based techniques, the method of statistically determined spatial drift correction (SDSDC) is normally employed.