-
Notifications
You must be signed in to change notification settings - Fork 23
/
Utils.py
342 lines (265 loc) · 12.6 KB
/
Utils.py
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
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
# Python program to print prime factors
import math
from sklearn import cluster
from skimage import data
import numpy as np
import cv2
import matplotlib.pyplot as plt
from scipy import ndimage
from scipy import signal
def factors_number(n):
# A function to print all prime factors of
# a given number n
n2 = int(n/2)
fn = []
for x in range(2,n2+1):
if n%x == 0:
if not([int(n/x), int(x)] in fn):
fn.append([int(x), int(n/x)]) #lower member of factor pair #upper member of factor pair
if fn == []: fn.append([1, n])
return fn
def rgb2gray(rgb):
# RGB array to gray
return np.dot(rgb[...,:3], [0.2989, 0.5870, 0.1140])
def km_clust(array, n_clusters):
## KMeans Segmentation Image
# Create a line array, the lazy way
if len(array.shape) == 3:
X = array.reshape((-1, array.shape[-1]))
else :
X = array.reshape((-1, 1))
# Define the k-means clustering problem
k_m = cluster.KMeans(n_clusters = n_clusters, n_init=4)
# Solve the k-means clustering problem
k_m.fit(X)
# Get the coordinates of the clusters centres as a 1D array
values = k_m.cluster_centers_.squeeze()
# Get the label of each point
labels = k_m.labels_
return(values, labels, k_m)
def kmeans_image(Z, n_clusters, weight_positional = 0):
#Kmeans in Image with dimensional position
#n_cluster is the number of n_clusters
#weight_positional is you can put weights to coordinates, i.e. Kmeans with two more dimensionas.
grid = np.indices((Z.shape[0], Z.shape[1]))
grid = grid.astype(float)
grid[0] = grid[0] / grid[0].max() * weight_positional
grid[1] = grid[1] / grid[1].max() * weight_positional
if len(Z.shape) == 3:
img = np.zeros((Z.shape[0], Z.shape[1], Z.shape[2] + 2))
img[:, :, :3] = Z#[:,:,1]
img[:, :, 3] = grid[0]
img[:, :, 4] = grid[1]
else:
img = np.zeros((Z.shape[0], Z.shape[1], 3))
img[:, :, 0] = Z#[:,:,1]
img[:, :, 1] = grid[0]
img[:, :, 2] = grid[1]
# Group similar grey levels using 8 clusters
values, labels, k_m = km_clust(img, n_clusters = n_clusters)
print (values.shape)
print (labels.shape)
# Create the segmented array from labels and values
img_segm = np.array([values[label][:3] for label in labels]) #np.choose(labels, values)
# Reshape the array as the original image
img_segm.shape = img[:,:,:3].shape
return img_segm
def order_points_rect(pts):
# initialzie a list of coordinates that will be ordered
# such that the first entry in the list is the top-left,
# the second entry is the top-right,
# the third is the bottom-right, and
#the fourth is the bottom-left
rect = np.zeros((4, 2), dtype = "float32")
# the top-left point will have the smallest sum, whereas
# the bottom-right point will have the largest sum
s = pts.sum(axis = 1)
rect[0] = pts[np.argmin(s)]
rect[2] = pts[np.argmax(s)]
# now, compute the difference between the points, the
# top-right point will have the smallest difference,
# whereas the bottom-left will have the largest difference
diff = np.diff(pts, axis = 1)
rect[1] = pts[np.argmin(diff)]
rect[3] = pts[np.argmax(diff)]
# return the ordered coordinates
return rect
def perspectiveTransform(Points):
#Transform cuadrilater image segmentation to rectangle image
# Return Matrix Transform
Points = np.array(Points)
Points_order = order_points_rect(Points)
#dst = np.asarray([[0, 0], [0, 1], [1, 1], [1, 0]], dtype = "float32")
(tl, tr, br, bl) = Points_order
# compute the width of the new image, which will be the
# maximum distance between bottom-right and bottom-left
# x-coordiates or the top-right and top-left x-coordinates
widthA = np.sqrt(((br[0] - bl[0]) ** 2) + ((br[1] - bl[1]) ** 2))
widthB = np.sqrt(((tr[0] - tl[0]) ** 2) + ((tr[1] - tl[1]) ** 2))
maxWidth = max(int(widthA), int(widthB))
# compute the height of the new image, which will be the
# maximum distance between the top-right and bottom-right
# y-coordinates or the top-left and bottom-left y-coordinates
heightA = np.sqrt(((tr[0] - br[0]) ** 2) + ((tr[1] - br[1]) ** 2))
heightB = np.sqrt(((tl[0] - bl[0]) ** 2) + ((tl[1] - bl[1]) ** 2))
maxHeight = max(int(heightA), int(heightB))
# now that we have the dimensions of the new image, construct
# the set of destination points to obtain a "birds eye view",
# (i.e. top-down view) of the image, again specifying points
# in the top-left, top-right, bottom-right, and bottom-left
# order
dst = np.array([
[0, 0],
[maxWidth , 0],
[maxWidth , maxHeight ],
[0, maxHeight ]], dtype = "float32")
M = cv2.getPerspectiveTransform(Points_order, dst)
return M, maxWidth, maxHeight
def subdivision_rect(factors, maxWidth, maxHeight, merge_percentaje = 0):
## From a rect (top-left, top-right, bottom-right, bottom-left) subidive in rectangle
#factors = factors_number(n_divide)[-1] # First factor is smaller
#if maxWidth > maxHeight:
# split_Width = [maxWidth / factors[1] * i for i in range(factors[1] + 1)]
# split_Height = [maxHeight / factors[0] * i for i in range(factors[0] + 1)]
#else:
# split_Width = [maxWidth / factors[0] * i for i in range(factors[0] + 1)]
# split_Height = [maxHeight / factors[1] * i for i in range(factors[1] + 1)]
merge_Width = maxWidth * merge_percentaje
merge_Height = maxHeight * merge_percentaje
split_Width = [maxWidth / factors[0] * i for i in range(factors[0] + 1)]
split_Height = [maxHeight / factors[1] * i for i in range(factors[1] + 1)]
sub_division = []
for j in range(len(split_Height) - 1):
for i in range(len(split_Width) - 1):
sub_division.append([(max(split_Width[i] - merge_Width, 0) , max(split_Height[j] - merge_Height , 0)),
(min(split_Width[i+1] + merge_Width , maxWidth - 1), max(split_Height[j] - merge_Height , 0)),
(min(split_Width[i+1] + merge_Width , maxWidth - 1), min(split_Height[j+1] + merge_Height, maxHeight - 1)),
(max(split_Width[i] - merge_Width, 0), min(split_Height[j+1] + merge_Height, maxHeight - 1))])
return np.array(sub_division)
def skeleton(bin_image, n_important = 100):
#From binary image (0,255) transform to skeleton edge
kernel_size = 3
edges = cv2.GaussianBlur((bin_image.copy()).astype(np.uint8),(kernel_size, kernel_size),0)
kernel = cv2.getStructuringElement(cv2.MORPH_CROSS,(5, 5))
height,width = edges.shape
skel = np.zeros([height,width],dtype=np.uint8) #[height,width,3]
temp_nonzero = np.count_nonzero(edges)
while (np.count_nonzero(edges) != 0 ):
eroded = cv2.erode(edges,kernel)
#cv2.imshow("eroded",eroded)
temp = cv2.dilate(eroded,kernel)
#cv2.imshow("dilate",temp)
temp = cv2.subtract(edges,temp)
skel = cv2.bitwise_or(skel,temp)
edges = eroded.copy()
"""This function returns the count of labels in a mask image."""
label_im, nb_labels = ndimage.label(skel)#, structure= np.ones((2,2))) ## Label each connect region
label_areas = np.bincount(label_im.ravel())[1:]
keys_max_areas = np.array(sorted(range(len(label_areas)), key=lambda k: label_areas[k], reverse = True)) + 1
keys_max_areas = keys_max_areas[:n_important]
L = np.zeros(label_im.shape)
for i in keys_max_areas:
L[label_im == i] = i
labels = np.unique(L)
label_im = np.searchsorted(labels, L)
return label_im>0
def angle_lines(skel_filter, n_important = 100, angle_resolution = 360, threshold = 100, min_line_length = 200, max_line_gap = 50, plot = False):
#Measure the angle of lines in skel_filter. Obs the angle is positive in clockwise.
rho = 1 # distance resolution in pixels of the Hough grid
theta = np.pi / angle_resolution # angular resolution in radians of the Hough grid
#threshold = 100 # minimum number of votes (intersections in Hough grid cell)
#min_line_length = 200 # minimum number of pixels making up a line
#max_line_gap = 50 # maximum gap in pixels between connectable line segments
lines = cv2.HoughLines(np.uint8(skel_filter),rho, theta, threshold)
lines_P = cv2.HoughLinesP(np.uint8(skel_filter),rho, theta, threshold, np.array([]) ,min_line_length, max_line_gap)
if lines_P is None:
print("linea no encontrada")
return 0
theta_P = [np.pi/2 + np.arctan2(line[0][3] - line[0][1],line[0][2]-line[0][0]) for line in lines_P[:n_important]]
theta = lines[0:n_important,0,1]
h = np.histogram(np.array(theta), bins = angle_resolution, range=(-np.pi,np.pi))
peaks = signal.find_peaks_cwt(h[0], widths= np.arange(2,4))
h_P = np.histogram(np.array(theta_P), bins = angle_resolution, range=(-np.pi,np.pi))
peaks_P = signal.find_peaks_cwt(h_P[0], widths= np.arange(2,4))
#h= np.histogram(np.array(theta), bins = angle_resolution, range=(-np.pi,np.pi))
#peaks = argrelextrema(h[0], np.greater)
#h_P = np.histogram(np.array(theta_P), bins = angle_resolution, range=(-np.pi,np.pi))
#peaks_P = argrelextrema(h_P[0], np.greater)
mesh = np.array(np.meshgrid(h[1][peaks], h_P[1][peaks_P]))
combinations = mesh.T.reshape(-1, 2)
index_min = np.argmin([abs(a-b) for a,b in combinations])
theta_prop = np.mean(combinations[index_min])
if plot:
print('Theta in HoughLines: ', h[1][peaks])
print('Theta in HoughLinesP: ', h_P[1][peaks_P])
print('combinations: ', combinations)
print('Theta prop: ', theta_prop)
Z1 = np.ones((skel_filter.shape))*255
Z2 = np.ones((skel_filter.shape))*255
for line in lines[0:n_important]:
rho,theta = line[0]
a = np.cos(theta)
b = np.sin(theta)
x0 = a*rho
y0 = b*rho
x1 = int(x0 + 1000*(-b))
y1 = int(y0 + 1000*(a))
x2 = int(x0 - 1000*(-b))
y2 = int(y0 - 1000*(a))
#print((x1,y1,x2,y2))
cv2.line(Z1,(x1,y1),(x2,y2),(0,0,255),2)
for line in lines_P[:n_important]:
x1,y1,x2,y2 = line[0]
cv2.line(Z2,(x1,y1),(x2,y2),(0,0,255),2)
plt.figure(0)
plt.figure(figsize=(16,8))
plt.imshow(skel_filter)
plt.title('Skel_filter')
fig, axs = plt.subplots(1, 2, figsize=(16,8))
axs[0].imshow(Z1)
axs[0].title.set_text('Lines HoughLines')
axs[1].imshow(Z2)
axs[1].title.set_text('Lines HoughLinesP')
fig, axs = plt.subplots(1, 2, figsize=(16,8))
axs[0].hist(lines[0:n_important,0,1], bins = 45, range=[-np.pi,np.pi])
axs[0].title.set_text('Lines HoughLines theta Histogram')
axs[1].hist(theta_P, bins = 45, range=[-np.pi,np.pi])
axs[1].title.set_text('Lines HoughLinesP theta Histogram')
#print(lines.shape)
#print(lines_P.shape)
return theta_prop
def rgb2hsv(rgb):
""" convert RGB to HSV color space
:param rgb: np.ndarray
:return: np.ndarray
"""
rgb = rgb.astype('float')
maxv = np.amax(rgb, axis=2)
maxc = np.argmax(rgb, axis=2)
minv = np.amin(rgb, axis=2)
minc = np.argmin(rgb, axis=2)
hsv = np.zeros(rgb.shape, dtype='float')
hsv[maxc == minc, 0] = np.zeros(hsv[maxc == minc, 0].shape)
hsv[maxc == 0, 0] = (((rgb[..., 1] - rgb[..., 2]) * 60.0 / (maxv - minv + np.spacing(1))) % 360.0)[maxc == 0]
hsv[maxc == 1, 0] = (((rgb[..., 2] - rgb[..., 0]) * 60.0 / (maxv - minv + np.spacing(1))) + 120.0)[maxc == 1]
hsv[maxc == 2, 0] = (((rgb[..., 0] - rgb[..., 1]) * 60.0 / (maxv - minv + np.spacing(1))) + 240.0)[maxc == 2]
hsv[maxv == 0, 1] = np.zeros(hsv[maxv == 0, 1].shape)
hsv[maxv != 0, 1] = (1 - minv / (maxv + np.spacing(1)))[maxv != 0]
hsv[..., 2] = maxv
return hsv
def lines2circles(List_lines, r_circle = 10, n_important = -1):
centers_filter = np.ones((0, 2))
centers = np.ones((0, 2))
for i,Poly in enumerate(List_lines[:n_important]):
P1 = np.mean(Poly[0:2], axis=0)
P2 = np.mean(Poly[2:4], axis=0)
distance = np.linalg.norm(P1-P2)
n_step = int(distance/(r_circle *2 + 1))
centers = np.concatenate((centers, np.array([P1 * t/n_step + P2 * (n_step-t)/n_step for t in range(n_step + 1)]).astype(int)))
centers = (np.unique(np.uint(centers), axis = 0)).astype(np.float)
### Filter circle so close
centers_filter = np.concatenate((centers_filter, np.array([centers[0]])))
for c in centers[1:]:
if np.min(np.linalg.norm(c - centers_filter, axis=1)) > 2 * r_circle:
centers_filter = np.concatenate((centers_filter,np.array([c])))
return np.uint(np.array(centers_filter))