# -*- python -*-
#
# Copyright INRIA - CIRAD - INRA
#
# Distributed under the Cecill-C License.
# See accompanying file LICENSE.txt or copy at
# http://www.cecill.info/licences/Licence_CeCILL-C_V1-en.html
#
# ==============================================================================
"""
A 3D 6-subiteration thinning algorithm for extracting medial lines
of Kalman Palagyi and Attila Kuba implementation
"""
# ==============================================================================
from __future__ import division, print_function, absolute_import
import numpy
# ==============================================================================
def _build_mask():
M1 = numpy.array(
[
[["x", "x", "0"], ["x", "x", "0"], ["x", "x", "0"]],
[["x", "x", "0"], ["1", "1", "0"], ["x", "x", "0"]],
[["x", "x", "0"], ["x", "x", "0"], ["x", "x", "0"]],
]
)
M2 = numpy.array(
[
[[".", ".", "0"], [".", ".", "0"], [".", ".", "0"]],
[[".", ".", "0"], ["1", "1", "0"], [".", ".", "0"]],
[[".", ".", "."], [".", "1", "."], [".", ".", "."]],
]
)
M3 = numpy.array(
[
[[".", ".", "0"], [".", ".", "0"], [".", ".", "."]],
[[".", ".", "0"], ["1", "1", "0"], [".", "1", "."]],
[[".", ".", "."], [".", "1", "."], [".", ".", "."]],
]
)
M4 = numpy.array(
[
[[".", ".", "0"], [".", ".", "0"], [".", ".", "0"]],
[[".", ".", "0"], ["1", "1", "0"], [".", ".", "0"]],
[[".", ".", "0"], [".", ".", "0"], [".", "1", "1"]],
]
)
M5 = numpy.array(
[
[["0", "0", "0"], ["0", "0", "0"], ["0", "0", "0"]],
[["x", "x", "0"], ["0", "1", "0"], ["x", "x", "0"]],
[["x", "x", "0"], ["1", "x", "0"], ["x", "x", "0"]],
]
)
M6 = numpy.array(
[
[["0", "0", "0"], ["0", "0", "0"], [".", ".", "0"]],
[["0", "0", "0"], ["0", "1", "0"], ["1", ".", "0"]],
[[".", ".", "0"], ["1", ".", "0"], [".", ".", "0"]],
]
)
U = [M1, M2, M3, M4, M5, M6]
D = []
for M in U:
D.append(M[:, :, ::-1])
W = []
N = []
E = []
S = []
for M in U:
tmp = numpy.zeros_like(M)
tmp[0, :, :] = numpy.rot90(M[0, :, :], 1)
tmp[1, :, :] = numpy.rot90(M[1, :, :], 1)
tmp[2, :, :] = numpy.rot90(M[2, :, :], 1)
W.append(tmp)
N.append(numpy.rot90(tmp))
E.append(numpy.rot90(tmp, 2))
S.append(numpy.rot90(tmp, 3))
return U, D, N, S, E, W
def _check_mask(T, M):
x_present = False
x_ok = False
for i in range(3):
for j in range(3):
for k in range(3):
if M[i, j, k] == ".":
continue
if M[i, j, k] == "0" and T[i, j, k] != 0:
return False
if M[i, j, k] == "1" and T[i, j, k] != 1:
return False
if M[i, j, k] == "x":
x_present = True
if T[i, j, k] == 1:
x_ok = True
if x_present:
if x_ok:
return True
return False
return True
def _applied_masks(mat, masks):
tmp = numpy.zeros_like(mat)
xx, yy, zz = numpy.where(mat == 1)
for i, xx_val in enumerate(xx):
x, y, z = xx_val, yy[i], zz[i]
block = mat[x - 1 : x + 2, y - 1 : y + 2, z - 1 : z + 2]
for mask in masks:
if _check_mask(block, mask):
tmp[x, y, z] = 1
return mat - tmp
[docs]
def skeletonize_thinning(img):
"""A 3D 6-subiteration thinning algorithm for extracting medial lines
of Kalman Palagyi and Attila Kuba implementation
Parameters
----------
img : ndarray, 3D
Returns
-------
skeleton : ndarray
The thinned image.
"""
mat_len_x, mat_len_y, mat_len_z = img.shape
mat_tmp = numpy.zeros((mat_len_x + 2, mat_len_y + 2, mat_len_z + 2), dtype=int)
mat_tmp[1:-1, 1:-1, 1:-1] = img.astype(int)
U, D, N, S, E, W = _build_mask()
# ==========================================================================
j = 0
while True:
j += 1
nb1 = numpy.count_nonzero(mat_tmp)
mat_tmp = _applied_masks(mat_tmp, U)
mat_tmp = _applied_masks(mat_tmp, D)
mat_tmp = _applied_masks(mat_tmp, N)
mat_tmp = _applied_masks(mat_tmp, S)
mat_tmp = _applied_masks(mat_tmp, E)
mat_tmp = _applied_masks(mat_tmp, W)
nb2 = numpy.count_nonzero(mat_tmp)
if nb1 == nb2:
break
# ==========================================================================
return mat_tmp[1:-1, 1:-1, 1:-1]
