Browse Prior Art Database

Color Coding Cluster Sizes From Electron Microscope Images

IP.com Disclosure Number: IPCOM000043788D
Original Publication Date: 1984-Sep-01
Included in the Prior Art Database: 2005-Feb-05
Document File: 1 page(s) / 12K

Publishing Venue

IBM

Related People

Krakow, W: AUTHOR

Abstract

Techniques are available for assigning particles of a given size from electron microscope images, a specific image density, and hence color. However, the number of objects analyzed is usually very limited (20 or less). Large numbers of particles can be color coded in descending order of size or some other particle feature (such as maximum cord length, x or y, axis projection, equivalent spherical diameter, etc.) if a cluster analysis is performed on a digitized image. In this technique, the electron microscope image is first digitized, after which a cluster analysis is performed on the digitized image. The particles are then sorted by size or particle feature, in a descending order via a computer program. An algorithm for grey levels, represented by the following expression 254 Sin(N-1)r/2 +1 is then applied.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 88% of the total text.

Page 1 of 1

Color Coding Cluster Sizes From Electron Microscope Images

Techniques are available for assigning particles of a given size from electron microscope images, a specific image density, and hence color. However, the number of objects analyzed is usually very limited (20 or less). Large numbers of particles can be color coded in descending order of size or some other particle feature (such as maximum cord length, x or y, axis projection, equivalent spherical diameter, etc.) if a cluster analysis is performed on a digitized image. In this technique, the electron microscope image is first digitized, after which a cluster analysis is performed on the digitized image. The particles are then sorted by size or particle feature, in a descending order via a computer program. An algorithm for grey levels, represented by the following expression 254 Sin(N- 1)r/2 +1 is then applied. In this expression, N is the particle number and r/2 is an adjustable parameter specifying the repeat cycle of the grey level coding. This function repeats when (N-1) r = f, and is unique over the ranges (N-1) r = nfT(n+1)f. By specifying the angle r, various repeat schemes can be achieved (e.g., r=90OE, N=5 - for this example, every fifth particle will have the same gray level). If very small angles are applied, the number of particles coded with different grey levels is increased. Up to 256 grey levels can be achieved. Once the grey levels have been assigned to each particle, look-up tables are lo...