1COOC(1)                     General Commands Manual                    COOC(1)
2
3
4

NAME

6       cooc,  cooc_features  - calculate the co-occurrence matrix and features
7       on it
8

SYNOPSIS

10       cooc image matrix xpos ypos xsize ysize dx dy flag
11
12       cooc_features matrix
13

DESCRIPTION

15       cooc creates a 256 by 256 one channel co-occurrence matrix of  the  box
16       determined  by  the  parameters (xp, yp; xs, ys) within the image file.
17       The matrix is written onto the Vasari image file matrix.  The displace‐
18       ment vector is determined by (dx, dy).  The user must ensure that there
19       is enough border pixels around the box within im dictated by  the  dis‐
20       placement vector (dx,dy) or else the program fails.  All entries of the
21       co-occurrence matrix are double  normalised  to  the  number  of  pairs
22       involved.   This function is a direct implementation of the paper: Har‐
23       alick R. M., Shanmugan K. and Dinstein I., 'Textural features for image
24       classification',  IEEE  Transactions  on Systems, Man, and Cybernetics,
25       Vol. SMC-3, No 6, Nov. 1973, pp 610-621.
26
27       If flag sym is 1, the created co-occurrence matrix is symmetric that is
28       dispacement  vectors  (dx,  dy),  (-dx,  -dy)  create  exactly the same
29       matrix.  If sym is 0, the created co-occurrence matrix is not symmetric
30       that  is  dispacement  vectors  (dx,  dy),  (-dx, -dy) create different
31       matrices.
32
33       Input image should be one band unsigned char image.
34
35       cooc_features calculates and prints at the standard error  output  fea‐
36       tures of the cooccurrence matrix matrix.
37

SEE ALSO

39       im_glds_matrix(3X), im_cooc_asm(3X), im_cooc_contrast(3X), im_cooc_cor‐
40       relation(3X), im_cooc_entropy(3X)
41

43       N. Dessipris
44

AUTHOR

46       N. Dessipris - 27/2/1992
47
48
49
50                                  27 Jan 1992                          COOC(1)
Impressum