Image analysis of electron microscope images using Boolean algebra operations

Abstract

For real time television image processing of electron microscope images it is inefficient to have image arithmetic performed on a host computer which requires considerable CPU and transfer time overhead. It is usually more efficient to use an analog preprocessor before image display such as the shading correctors used for particle analysis. These techniques can also be applied to digital TV frame stores which have simple arithmetic units as preprocessors to a digital frame store. In this case sampled images are either added, subtracted or divided much as is done in many STEM type instruments at slower scan rates. Alternatively, the digital frame store can have an arithmetic and Boolean algebra processor capability which operates on digitized images in the frame store in real time. Such processors are available commercially and one has been adapted for use on various electron microscopes at the IBM Research Center.