Characteristics of Different Frequency Ranges in Scanning Electron Microscope Images

We demonstrate a new approach to characterize the frequency range in general scanning electron microscope (SEM) images. First, pure frequency images are generated from low frequency to high frequency, and then, the magnification of each type of frequency image is implemented. By comparing the edge percentage of the SEM image to the self-generated frequency images, we can define the frequency ranges of the SEM images. Characterization of frequency ranges of SEM images benefits further processing and analysis of those SEM images, such as in noise filtering and contrast enhancement.