1LOG_MASKS(3) Library Functions Manual LOG_MASKS(3)
2
3
4
6 im_log_dmask, im_log_imask - create a laplacian of gaussian (log) DOU‐
7 BLEMASK or INTMASK
8
10 #include <vips/vips.h>
11
12
13 DOUBLEMASK *im_log_dmask( name, sigma, min_amplitude )
14 char *name;
15 double sigma, min_amplitude;
16
17 INTMASK *im_log_imask( name, sigma, min_amplitude )
18 char *name;
19 double sigma, min_amplitude;
20
22 Both functions create a circularly symmetric laplacian of Gaussian
23 mask. The size of the mask is determined by the variable min_ampli‐
24 tude; if for instance the value .1 is entered this means that the pro‐
25 duced mask is clipped at values within 10 persent of zero, and where
26 the change between mask elements is less than 10%. The mask can be
27 directly used with the vasari convolution programs, the default offset
28 set is 0.
29
30 The program uses the following equation: (from Handbook of Pattern
31 Recognition and image processing by Young and Fu, AP 1986 pages
32 220-221):
33
34 H(r) = (1 / (2 * M_PI * s4)) *
35 (2 - (r2 / s2)) * exp(-r2 / (2 * s2))
36
37 where s2 = sigma * sigma, s4=s2 * s2, r2 = r * r. The generated mask
38 has odd size and its maximum value is normalised to either 100
39 (log_imask) or to 1.0 (log_dmask)
40
41 im_log_dmask() creates a DOUBLEMASK laplacian of Gaussian mask with
42 maximum value normalised to 1.0.
43
44 im_log_imask() creates a INTMASK laplacian of Gaussian mask with maxi‐
45 mum value normalised to 100.
46
48 The functions return NULL on erorr.
49
51 im_gauss_dmask(3), im_conv(3)
52
54 N. Dessipris
55
57 N. Dessipris - 06/12/1991
58
59
60
61 6 December 1991 LOG_MASKS(3)