IM_SGLDS_MATRIX

Section: C Library Functions (3)
Updated: 10 May 1991
Index Return to Main Contents

NAME

im_glds_matrix, im_glds_asm, im_glds_contrast, im_glds_mean, im_glds_entropy - calculate the spatial grey level difference matrix and features on it

SYNOPSIS

#include <vips/vips.h>

int im_glds_matrix(im, m, xp, yp, xs, ys, dx, dy)
IMAGE *im, *m;
int xp, yp, xs, ys;
int dx, dy;

int im_glds_asm(m, asmoment)
IMAGE *m;
double *asmoment;

int im_glds_contrast(m, contrast)
IMAGE *m;
double *contrast;

int im_glds_entropy(m, entropy)
IMAGE *m;
double *entropy;

int im_glds_mean(m, mean)
IMAGE *m;
double *mean;

DESCRIPTION

im_glds_matrix() creates a 256 by 1 one channel spatial grey level difference matrix (sglds) of the box determined by the parameters (xp, yp; xs, ys) within the image pointed by the IMAGE descriptor im. The matrix is written onto the IMAGE descriptor m. The displacement vector is determined by (dx, dy). The user must ensure that there is enough border pixels around the box within im dictated by the displacement vector (dx,dy) or else the program fails. im should be one-band unsigned char.

All entries of the sgld matrix are double normalised to the number of pairs involved. This function is a direct implementation of the paper: Haralick R. M., Shanmugan K. and Dinstein I., 'Textural features for image classification', IEEE Transactions on Systems, Man, and Cybernetics, Vol. SMC-3, No 6, Nov. 1973, pp 610-621.

im_glds_asm() calculates the angular second moment of the co-occurrence matrix held by m. The result is returned into the location pointed by asmoment.

im_glds_contrast() calculates the contrast of the sglds matrix held by m. The result is returned into the location pointed by contrast.

im_glds_entropy() calculates the entropy of the sglds matrix held by m. The result is returned into the location pointed by entropy.

im_glds_mean() calculates the mean of the sglds matrix held by m. The result is returned into the location pointed by mean.

RETURNED VALUES

All functions returns 0 on success and -1 on error.

COPYRIGHT

N. Dessipris

AUTHOR

N. Dessipris - 10/05/1991