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
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
|
import cv2
import argparse
import datetime
import numpy as np
from matplotlib import gridspec
from matplotlib import pyplot as plt
btn_down = False
SCALE = 9.2661 # pixel per micron
RED = 'mcherry'
GREEN = 'GFP'
BLUE = 'DAPI'
parser = argparse.ArgumentParser(usage="drawCoLoc -f RGB_image_file_path [<-b> <--noPCC> <-n> <-r resolution>]",
description="\n\nHere's a script for co-localization analysis (line profile) on RGB image")
parser.add_argument('-f', metavar='file path', help='the path of the RGB image', dest="fpath", type=str, required=True)
parser.add_argument('-b', '--blue', help='add this parameter if you need blue channel', default=False, action='store_true')
parser.add_argument('--noPCC', help='add this parameter if you do not need label Person Cor', default=False, action='store_true')
parser.add_argument('-n', '--normalization', help='add this parameter if you need normalization for each channel', default=False, action='store_true')
parser.add_argument('-r', help='resolution if you need modify default values 9.2661 for dv usual use', dest='resolution', type=float, default=SCALE)
def get_points(im):
# Set up data to send to mouse handler
data = {}
data['im'] = im.copy()
data['lines'] = []
# Set the callback function for any mouse event
cv2.imshow("Image", im)
cv2.setMouseCallback("Image", mouse_handler, data)
cv2.waitKey(0)
# Convert array to np.array in shape n,2,2
points = np.uint16(data['lines'])
draw_coloc(im, data['im'], points)
# return points, data['im']
return True
def mouse_handler(event, x, y, flags, data):
global btn_down
if event == cv2.EVENT_LBUTTONUP and btn_down:
# if you release the button, finish the line
btn_down = False
data['lines'][0].append((x, y))
cv2.line(data['im'], data['lines'][0][0], data['lines'][0][1], (255, 255, 255), 2)
cv2.imshow("Image", data['im'])
elif event == cv2.EVENT_MOUSEMOVE and btn_down:
# thi is just for a ine visualization
image = data['im'].copy()
cv2.line(image, data['lines'][0][0], (x, y), (255, 255, 255), 1)
cv2.imshow("Image", image)
elif event == cv2.EVENT_LBUTTONDOWN and len(data['lines']) < 2:
btn_down = True
data['lines'].insert(0, [(x, y)]) # prepend the point
cv2.imshow("Image", data['im'])
def normalization(x):
_range = np.max(x) - np.min(x)
return (x - np.min(x)) / _range
def get_line(img, pts):
c = []
for p in pts:
dx, dy = int(p[-1, 0]) - int(p[0, 0]), int(p[-1, 1]) - int(p[0, 1])
if dx >= dy:
k, d = dy / dx, ((dx * dx + dy * dy) ** (1 / 2)) / args.resolution
for t in range(dx):
i = int(p[0, 0]) + t
j = int(p[0, 1]) + round(k * t)
c.append(img[j, i, ::-1])
ds = np.linspace(0, d, dx)
c = np.array(c).T
else:
k, d = dx / dy, ((dx * dx + dy * dy) ** (1 / 2)) / args.resolution
for t in range(dy):
i = int(p[0, 0]) + round(k * t)
j = int(p[0, 1]) + t
c.append(img[j, i, ::-1])
ds = np.linspace(0, d, dy)
c = np.array(c).T
if args.normalization:
c = [normalization(x) for x in c]
return ds, c
def draw_coloc(img, final_image, pts):
ds, c = get_line(img, pts)
cor = np.corrcoef(c[0], c[1])
fig = plt.figure(figsize=(18, 5))
gs = gridspec.GridSpec(1, 2, width_ratios=[1, 2])
ax0 = plt.subplot(gs[0])
ax0.imshow(final_image[:, :, ::-1]);
ax0.set_xticks([]);
ax0.set_yticks([])
ax1 = plt.subplot(gs[1])
ax1.plot(ds, c[0], color='red', label=RED)
ax1.plot(ds, c[1], color='green', label=GREEN)
if args.blue:
ax1.plot(ds, c[2], color='blue', label=BLUE)
if args.noPCC:
print(f"Person's corr={cor[0][1]:.2f}")
else:
ax1.text(0.85, 0.92, f"Person's corr={cor[0][1]:.2f}", fontsize=12, transform=ax1.transAxes)
# plt.legend(loc='lower center')
ax1.set_ylabel('Intensity');
ax1.set_yticks([])
ax1.set_xlabel('Distance (micron)')
dt = datetime.datetime.now().strftime('%Y%m%d %H%M%S')
plt.tight_layout()
plt.savefig(f'.\\{"".join(args.fpath.split(".")[:-1])}-{dt}.jpg')
plt.show()
if __name__ == '__main__':
args = parser.parse_args()
img = cv2.imread(args.fpath, 1)
if get_points(img):
print('Succeed!')
|
