Article
An image processing method for scanning electron microscopy based on the information transmission theory
Article first published online: 4 FEB 2005
DOI: 10.1002/jemt.1060020202
Copyright © 1985 Wiley-Liss, Inc.
Additional Information
How to Cite
Kanaya, K., Oho, E., Naka, M., Koyanagi, T. and Sasaki, T. (1985), An image processing method for scanning electron microscopy based on the information transmission theory. J. Elec. Microsc. Tech., 2: 73–87. doi: 10.1002/jemt.1060020202
Publication History
- Issue published online: 4 FEB 2005
- Article first published online: 4 FEB 2005
- Manuscript Accepted: 9 JUL 1984
- Manuscript Received: 30 MAR 1984
- Abstract
- References
- Cited By
Keywords:
- Response function;
- Gradient;
- Acutance;
- Laplacian;
- Edge sharpening
Abstract
Based on the information transmission theory, topographic image signals in scanning electron microscopy are used to evaluate contrast, gradient, acutance, and Laplacian operator, the total of which represent the image sharpness of an edge line.
One may consider the impulse and step functions as an input to the Gaussian system function of a low-pass filter, the impulse and step response functions possibly representing a single spot and image contrast of an edge profile, respectively. It is shown that the response function of acutance defined as the power of the gradient normalized by density is a more realistic representation of image edge sharpness. Also, edge sharpness can be greatly enhanced by utilizing the Laplacian operator through digital image processing for a disk specimen model with a rounded edge.
Contrast increased by specimen tilt, and an edge effect due to side-scattered electrons, as well as the signal attenuation by specimen collection, are consistently obtained as the response function in the system.
The exact measurement of spot size and edge-to-edge resolution, and image sharpness improvement, are derived by digital image processing.

1097-0029/asset/JEMT_left.gif?v=1&s=328bfc988dc6023d3c0e23054e952bf3444cffc8)
1097-0029/asset/cover.gif?v=1&s=43ac34a0c5790efa9b3c9852600b3b1d1fb6e338)