bcosfire_python/BCOSFIRE.py at main · vicimikul/bcosfire_python · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
#!/usr/bin/env python

""" Python implementation of the B-COSFIRE algorithm.

B-COSFIRE aims to detect elongated patterns in images such as blood vessels in retinal images.
It uses the existing COSFIRE (Combination Of Shifted Filter Responses) algorithm to get orientation selectivity.
More specifically, the algorithm computes the weighted geometric mean of the output of a pool of Difference-of-Gaussians filters to detect vessel-like patterns in images.

This program is free software: you can redistribute it and/or modify it under
the terms of the BSD General Public License as published by The COSFIRE Consolidation Project, version 0.0.1.

This program is distributed in the hope that it will be useful, but WITHOUT
ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
FOR A PARTICULAR PURPOSE. See the BSD General Public License for more details.
You should have received a copy of the BSD General Public License along with
this program. If not, see https://github.com/Brains-for-hire/bcosfire_python/blob/main/LICENSE.
"""


import numpy as np
import cosfire as c

def BCOSFIRE(img_rgb, mask=[]):
	## Model configuration

	proto_symm = np.zeros(shape=(201,201)).astype(np.uint8)
	proto_symm[:,100] = 255

	subject = 255 - img_rgb[:,:,1]
	subject = subject/255

	cx, cy = (100,100)

	# Symmetrical filter
	cosfire_symm = c.COSFIRE(
		c.CircleStrategy(c.DoGFilter, (2.4, 1), prototype=proto_symm, center=(cx,cy), rhoList=range(0,9,2), sigma0=3,  alpha=0.7,
		rotationInvariance = np.arange(12)/12*np.pi)
	   ).fit()
	resp_symm = cosfire_symm.transform(subject)

	# Asymmetrical filter
	cosfire_asymm = c.COSFIRE(
			c.CircleStrategy(c.DoGFilter, (1.8, 1), prototype=proto_symm, center=(cx,cy), rhoList=range(0,23,2), sigma0=2,  alpha=0.1,
			rotationInvariance = np.arange(24)/12*np.pi)
		   ).fit()


	# Make asymmetrical
	asymmTuples = []
	for tupl in cosfire_asymm.strategy.tuples:
		if tupl[1] <= np.pi:
			asymmTuples.append(tupl)
	cosfire_asymm.strategy.tuples = asymmTuples
	resp_asymm = cosfire_asymm.transform(subject)


	resp = resp_symm + resp_asymm
	resp_symm = c.rescaleImage(resp_symm, 0, 255)
	resp_asymm = c.rescaleImage(resp_asymm, 0, 255)
	resp = np.multiply(resp, mask)
	resp = c.rescaleImage(resp, 0, 255)
	segresp = np.where(resp > 37, 255, 0)
	return resp,segresp


if __name__ == '__main__':
	import sys
	import cv2
	import matplotlib.pyplot as plt
	img_rgb = cv2.cvtColor(cv2.imread(sys.argv[1],1), cv2.COLOR_BGR2RGB)
	img_rgb = np.pad(img_rgb,((20,20),(20,20),(0,0))) # Add some padding to get rid of the white edges in the output vessel image
	mask = np.ones(shape=img_rgb.shape[:-1])

	resp,segresp = BCOSFIRE(img_rgb,mask)
	resp = resp[20:-20,20:-20] # Remove padding
	p_vessel = resp/np.amax(resp)
	# plt.imsave('./figures/sample_0_out.png',p_vessel,cmap='gray')
	plt.imshow(p_vessel,cmap='gray')
	plt.show()


__author__ = "Murat Cankaya"
__contact__ = "info@brainsforhire.eu"
__copyright__ = "Copyright 2020, Brains for hire UG"
__credits__ = ["Nicola Strisciuglio"]
__date__ = "2020/12/14"
__deprecated__ = False
__email__ =  "info@"
__license__ = "BSD3"
__version__ = "0.0.1"