Using this facility in DigImage, the fractal dimension (Kolmogorov capacity) of a selected intensity contour may be evaluated using a line counting algorithm. With this technique the image a line throught the image is divided ino into a number of equal sized sections. The number of sections containing a crossing of the required intensity contour is counted. The process is then repeated with a different size of section. The fractal dimension of the contour is related to the slope of the relationship between the logarithm of the number of sections through which the contour passes and the logarithm of the size of the sections. A linear relationship between these two is an indication of self-similarity, the key feature of fractal contours. As an option, DigImage is able to produce a plot of this information, along with estimates of the slope of the relationship. By producing such a plot, the degree to which a contour is fractal may be visualised.
DigImage performs this calculation for both horizontal and vertical lines falling within the specified region. All horizontal lines within the region are used for the horizontal dimension evaluation, and all vertical lines for the corresponding vertical line calculation.
Note that this technique is very similar to [F Fractal dimension - 2D box counting].
RETURN VALUES DigImage fits a power law to the data using a least-squares fit in log-log space. This fit is plotted along with the raw data, and the fit printed out. The coefficients fitted are also available through the !!Rn mechanism:
Vertical fractal (plotted first)
!!R3 The scale factor for the fit.
!!R4 The power law - this is the "fractal dimension"
!!R5 The root mean square error in the fit.
Horizontal fractal (plotted second)
!!R0 The scale factor for the fit.
!!R1 The power law - this is the "fractal dimension"
!!R2 The root mean square error in the fit.
Note that you will normally require a E, F or G modifier to return the floating point representation of these return variables (e.g. !!GR1).
Encapsulated PostScript (.EPS) and Data Only PostScript (.DPS) can not be printed directly, except using DigiFile. Normally they would be included within some other package (eg. a word processor) to be printed. The .DPS file is a special version of .EPS which does not contain the normal DigImage\PostScr\Header.PS and DigImage\PostScr\GraphVDU.PS header files. These may need to be added to the PostScript prolog used by the application into which the .DPS file is inserted.
If no extension is specified, then the default PostScript type specified in [;P Printers] in CONFIGUR.EXE will be produced, and the appropriate extension added.
This file may be sent to the printer immediately by specifying PRN as the file name.
{If PostScript file specified} Default, Line, Greyscale or Colour printing ? DigImage is able to produce either grey-scale or colour PostScript files. The Default option is controlled by the setting in the Configur.EXE utility. If colour is selected, then the file MUST be printed on a level 2 (or higher) PostScript printer.
Default Colour type specified in Configur.EXE
Line Greyscale images and dashed line types
Greyscale Greyscale images and greyscale lines
Colour Colour images and colour lines - requires
Level 2 PostScript