-
Notifications
You must be signed in to change notification settings - Fork 1
/
main.py
245 lines (171 loc) · 4.2 KB
/
main.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
import sys
import cv2
import numpy as np
photo = []
tokens = []
gaps = []
model = []
def read():
global photo
path = ""
args = sys.argv[1:]
if len(args) >= 2:
path = args[1]
photo = cv2.imread(path, cv2.IMREAD_GRAYSCALE)
return
def delete_outliers(arr):
res = []
m, s = np.array(arr).mean(), np.array(arr).std()
for i in arr:
if abs(i - m) <= s:
res.append(i)
return res
def find_height_threshold(photo):
gaps = []
l, r = photo.shape
for i in range(r):
e = 0
for j in range(l):
if photo[j][i] == 255:
e = e + 1
else:
if e != 0 and e != j:
gaps.append(e)
e = 0
gaps = delete_outliers(gaps)
if not gaps:
return l
return np.array(gaps).mean()
def find_row_threshold(photo):
gaps = []
l, r = photo.shape
for i in range(l):
e = 0
for j in range(r):
if photo[i][j] == 255:
e = e + 1
else:
if e != 0 and e != j:
gaps.append(e)
e = 0
gaps = delete_outliers(gaps)
if not gaps:
return l
return np.array(gaps).mean()
def delete_first_row(photo):
_, r = photo.shape
for p in range(r):
if photo[0][p] != 255:
return False
return True
def delete_last_row(photo):
l, r = photo.shape
for p in range(r):
if photo[l - 1][p] != 255:
return False
return True
def delete_first_column(photo):
l, _ = photo.shape
for p in range(l):
if photo[p][0] != 255:
return False
return True
def delete_last_column(photo):
l, r = photo.shape
for p in range(l):
if photo[p][r - 1] != 255:
return False
return True
def normalize(photo):
photo[photo > 127] = 255
while delete_first_row(photo):
photo = photo[1:, :]
while delete_last_row(photo):
photo = photo[:-1, :]
while delete_first_column(photo):
photo = photo[:, 1:]
while delete_last_column(photo):
photo = photo[:, :-1]
return photo
def find_line_height(photo):
height = 0
photo = normalize(photo)
height_threshold = find_height_threshold(photo)
l, r = photo.shape
for i in range(r):
e = 0
for k in range(l):
if photo[k][i] == 255:
e = e + 1
if e > height_threshold:
height = max(k, height)
break
else:
e = 0
if e <= height_threshold:
return l
if height < height_threshold:
height = l
return height
def add_row_tokens(photo):
global tokens
last_block = 0
e = 0
photo = normalize(photo)
row_threshold = find_row_threshold(photo)
l, r = photo.shape
for i in range(r):
has_black = False
for j in range(l):
if photo[j][i] != 255:
has_black = True
break
if not has_black:
e = e + 1
else:
if e > row_threshold:
tokens.append(photo[:, last_block:i + 1])
last_block = i + 1
e = 0
if last_block < r:
tokens.append(photo[:, last_block:])
return
def tokenize():
global photo, tokens
while len(photo) != 0:
photo = normalize(photo)
lh = find_line_height(np.array(photo))
cut = photo[:lh + 1, :]
add_row_tokens(cut)
photo = photo[lh + 1:, :]
return
def is_white(token_photo):
l, r = token_photo.shape
for i in range(l):
for j in range(r):
if token_photo[i][j] != 255:
return False
return True
def filter_tokens():
global tokens
filtered_tokens = []
for i in tokens:
if not is_white(i):
filtered_tokens.append(i)
tokens = filtered_tokens
return
def train():
global tokens
for i in tokens:
cv2.imshow("c", i)
cv2.waitKey(0)
return
def find_Answer():
return
def run():
read()
tokenize()
filter_tokens()
train()
find_Answer()
run()