[mmdist] [Up] [mmopentransf] | Image Transforms |
Implemented in Python.
f | Image Binary image. |
g | Image Binary image.
Marker image |
Bc | Structuring Element (metric for distance). Default:
|
METRIC | String 'EUCLIDEAN' if specified. Default:
|
y | Image uint16 (distance image). |
mmgdist creates the geodesic distance image
y
of the binary image
f
relative to the binary image
g
. The value of
y
at the pixel
x
is the length of the smallest path between
x
and
f
. The distances available are based on the Euclidean metrics and on metrics generated by a neighbourhood graph, that is characterized by a connectivity rule defined by the structuring element
Bc
. The connectivity for defining the paths is consistent with the metrics adopted to measure their length. In the case of the Euclidean distance, the space is considered continuos and, in the other cases, the connectivity is the one defined by
Bc
.
>>> f=mmbinary([ [1,1,1,1,1,1], [1,1,1,0,0,1], [1,0,1,0,0,1], [1,0,1,1,0,0], [0,0,1,1,1,1], [0,0,0,1,1,1]])
>>> g=mmbinary([ [0,0,0,0,0,0], [1,1,0,0,0,0], [0,0,0,0,0,0], [0,0,0,0,0,0], [0,0,0,0,0,0], [0,0,0,0,0,1]])
>>> y=mmgdist(f,g,mmsecross())
>>> print y
[[ 1 1 2 3 4 5] [ 0 0 1 65535 65535 6] [ 1 65535 2 65535 65535 7] [ 2 65535 3 4 65535 65535] [65535 65535 4 3 2 1] [65535 65535 65535 2 1 0]]
To generate useful Distance transforms, the structuring elements must be symmetric and with the origin included.
The Euclidean Distance transform is rounded to the nearest integer, since it is represented in an unsigned integer image. You should use the
mmsebox
structuring element when computing the Euclidean Distance transform.
def mmgdist(f, g, Bc=None, METRIC=None): if Bc is None: Bc = mmsecross() assert METRIC is None,'Does not support EUCLIDEAN' fneg,gneg = mmneg(f),mmneg(g) y = mmgray(gneg,'uint16',1) ero = mmintersec(y,0) aux = y i = 1 while not mmisequal(ero,aux): aux = ero ero = mmcero(gneg,fneg,Bc,i) y = mmaddm(y,mmgray(ero,'uint16',1)) i = i + 1 y = mmunion(y,mmgray(ero,'uint16')) return y
[mmdist] [Up] [mmopentransf] | ![]() |
Copyright (c) 2003, Roberto A. Lotufo, UNICAMP-University of Campinas; Rubens C. Machado, CenPRA-Renato Archer Research Center. |