The probability mass function of regular difference (PMFRD) represents the
probabilities of a random variable W, being $W=(i-j)$.
$Pr(w+256)=P((i-j)=w)$, $-255 \leq w \leq 255$
Where i is one intensity id line in COM matrix and j is one intensity id
column in COM. Thus $Pr(w+256)=P((i-j)=w)$ is the probability of happen an
intensity jump of value i to value $j=i+w$, with -255<=w<=255. This function
calculates a similar difference probability proposed in [1]. The COM matrix
need have 256x256 elements.
References:
[1] R.M. Haralick, K. Shanmugam, and Its' Hak Dinstein. 'Textural Features
for Image Classification'. In: Systems, Man and Cybernetics, IEEE
Transactions on SMC-3.6 (Nov. 1973), pages 610-621. ISSN: 0018-9472.
After starting the main routine just type the following command at the
prompt:
Pr = pmfrd(COM);
Input:
COM is the Co-Occurrence matrix. A 2D matrix with 256 lines and 256 columns.
Output:
Pr is the probability mass function, where $Pr(w+256)=P((i-j)=w)$ is the
probability of happen an intensity jump of value i to value j=i+w, with
-255<=w<=255. Pr is a vector with 511 elements.
W [OPTIONAL] is a vector with the intensities jumps $(i-j)=w$.
W=[-255:255];
For help, bug reports and feature suggestions, please visit:
http://nongnu.org/bsltl/
Package: bsltl
