Mercurial > repos > imgteam > binary2labelimage

comparison 2d_split_binaryimage_by_watershed.py @ 5:7f8102bdbfa1 draft default tip

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
planemo upload for repository https://github.com/BMCV/galaxy-image-analysis/tree/master/tools/binary2labelimage/ commit 48df7d9c58fb88e472caeb4d4a1e14170d79b643
author imgteam
date Mon, 12 May 2025 08:15:44 +0000
parents 984358e43242
children
comparison
equal deleted inserted replaced
4:984358e43242 5:7f8102bdbfa1
1 import argparse 1 import argparse
2 import sys 2 import sys
3 3
4 import numpy as np
4 import skimage.io 5 import skimage.io
5 import skimage.util 6 import skimage.util
6 from scipy import ndimage as ndi 7 from scipy import ndimage as ndi
7 from skimage.feature import peak_local_max 8 from skimage.feature import peak_local_max
8 from skimage.morphology import watershed 9 from skimage.segmentation import watershed
9 10
10 11
11 if __name__ == "__main__": 12 if __name__ == "__main__":
12 parser = argparse.ArgumentParser(description='Split binaryimage by watershed') 13 parser = argparse.ArgumentParser(description='Split binaryimage by watershed')
13 parser.add_argument('input_file', type=argparse.FileType('r'), default=sys.stdin, help='input file') 14 parser.add_argument('input_file', type=argparse.FileType('r'), default=sys.stdin, help='input file')
15 parser.add_argument('min_distance', type=int, default=100, help='Minimum distance to next object') 16 parser.add_argument('min_distance', type=int, default=100, help='Minimum distance to next object')
16 args = parser.parse_args() 17 args = parser.parse_args()
17 18
18 img_in = skimage.io.imread(args.input_file.name) 19 img_in = skimage.io.imread(args.input_file.name)
19 distance = ndi.distance_transform_edt(img_in) 20 distance = ndi.distance_transform_edt(img_in)
20 local_maxi = peak_local_max(distance, 21
21 indices=False, 22 local_max_indices = peak_local_max(
22 min_distance=args.min_distance, 23 distance,
23 labels=img_in) 24 min_distance=args.min_distance,
24 markers = ndi.label(local_maxi)[0] 25 labels=img_in,
26 )
27 local_max_mask = np.zeros(img_in.shape, dtype=bool)
28 local_max_mask[tuple(local_max_indices.T)] = True
29 markers = ndi.label(local_max_mask)[0]
25 res = watershed(-distance, markers, mask=img_in) 30 res = watershed(-distance, markers, mask=img_in)
26 31
27 res = skimage.util.img_as_uint(res) 32 res = skimage.util.img_as_uint(res)
28 skimage.io.imsave(args.out_file.name, res, plugin="tifffile") 33 skimage.io.imsave(args.out_file.name, res, plugin="tifffile")