-
Notifications
You must be signed in to change notification settings - Fork 33
/
Copy pathchinese_out_jiguan.py
630 lines (583 loc) · 26.5 KB
/
chinese_out_jiguan.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
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
#coding:utf-8
import cv2 as cv
import numpy as np
import math
import matplotlib.pyplot as plt
from PIL import Image
import ABO_mix
import chinese_ocr
def chinese_o(th2, shrinkTwoTimesTranslation_copy, x, y, w, h, cha_num, get_all, is_ABO):
return_change = 0
chinese_num_lim = 0
cut = ~th2[y:y+h, x:x+w]#从表格中分割出目标矩形框
copy_cut = ~th2[y:y+h, x:x+w]
cv.imwrite('cut_ori.jpg', cut)
##### 水平投影分割 #####
b = [0 for z in range(0, h)]
for j in range(0,h):
for i in range(0,w):
if copy_cut[j,i]==0:
b[j]+=1
copy_cut[j,i]=255
print b
b_total = 0
for j in range(2,h-2):
b_total = b_total + b[j]
blank_counting = 0
for j in range(3,h-3):
for i in range(3,w-3):
if cut[j,i]==0:
blank_counting+=1
print "blank_counting/w:", blank_counting/w
if float(blank_counting)/w <= 0.5:#排除空框
return return_change, '无'
else:
print sum(b)/(8*h) #4
word_start = 0
word_b_set = []
if sum(b)/(8*h) <= 3:#设定水平分割时的最低阈值
standard = 3
else:
standard = sum(b)/(8*h)#4
for j in range(2,h-2):
if b[j] >= standard and word_start == 0:
word_start_b = j
word_start = 1
#if (b[j] < standard or j == h-3) and word_start == 1:
if (b_total < standard or j == h-3) and word_start == 1:
word_end_b = j
word_start = 0
if abs(word_end_b - word_start_b) >= 10:#行像素超过10则判断为一行手写字
word_b = (word_start_b, word_end_b)
word_b_set.append(word_b)
b_total = b_total - b[j]
print word_b_set
for j in range(0,h):
for i in range(0,b[j]):
copy_cut[j,i]=0
cv.imwrite('copy_cut.jpg', copy_cut)
##### 垂直投影分割 #####
word_set = []
word_cluster_cut_set = []
word_cluster_color_cut_set = []
word_count = 0
for row_cut in range(0,len(word_b_set)):
cut = ~th2[y + word_b_set[row_cut][0]:y + word_b_set[row_cut][1], x:x+w]
h = abs(word_b_set[row_cut][1] - word_b_set[row_cut][0])
a = [0 for z in range(0, w)]
#print(a) #a = [0,0,0,0,0,0,0,0,0,0,...,0,0]初始化一个长度为w的数组，用于记录每一列的黑点个数
#记录每一列的波峰
for j in range(0,w): #遍历一列
for i in range(0,h): #遍历一行
if cut[i,j]==0: #如果该点为黑点
a[j]+=1 #该列的计数器加一计数
cut[i,j]=255 #记录完后将其变为白色
print a
print sum(a)/(2*w)#2
if sum(a)/(2*w) <= 3:#设定垂直分割时的最低阈值
standard = 3#3
else:
standard = sum(a)/(2*w)#2
word_cutting = 0
word_cut_counting = 0
word_cut_counting_over = 0
word_total_length = 0
word_start_a = 3#避免受边框的干扰
for j in range(3,w-2):#垂直分割
if a[j] >= standard and word_start == 0:
word_start_a = j
word_start = 1
#连续三个像素都低于阈值则设置分割点
if word_cutting == 1 and a[j] < standard and a[j-1] < standard and a[j-2] < standard:
word_cut_end = j
word_cut_counting+=1
if a[j] < standard and a[j-1] < standard and a[j-2] < standard and word_start == 1 and word_cutting == 0 and word_cut_counting_over == 0:
word_cut_begin = j-2
word_cut_end = j
word_cutting = 1
word_cut_counting = 1
'''
if word_cutting == 1 and a[j] < standard and a[j-1] < standard:
word_cut_end = j
word_cut_counting+=1
if a[j] < standard and a[j-1] < standard and word_start == 1 and word_cutting == 0 and word_cut_counting_over == 0:
word_cut_begin = j-1
word_cut_end = j
word_cutting = 1
word_cut_counting = 1
'''
if (a[j] >= standard and (word_cutting == 1 or word_cut_counting_over == 1)) or (word_cut_counting >= cha_num and word_cutting == 1) or (j == w-3 and word_cut_counting_over == 0):
if word_cut_counting >= cha_num:
word_cut_counting_over = 1
else:
word_cut_counting_over = 0
word_cutting = 0
word_cut_counting = 0
if j == w-3 and word_cut_counting_over == 0:
word_end_a = j
else:
word_end_a = (word_cut_begin + word_cut_end)//2
if abs(word_end_a - word_start_a) >= 12:#列像素超过12则判断为一个手写字
word = (word_start_a, word_end_a, word_b_set[row_cut][0], word_b_set[row_cut][1])
word_cut = ~th2[y + word_b_set[row_cut][0]:y + word_b_set[row_cut][1], x + word_start_a:x + word_end_a]
word_cut_color = shrinkTwoTimesTranslation_copy[y + word_b_set[row_cut][0]:y + word_b_set[row_cut][1], x + word_start_a:x + word_end_a]
word_cut_mean = cv.mean(word_cut)
print "word_cut_mean[0]:", word_cut_mean[0]
if word_cut_mean[0] < 245:#排除空框
word_set.append(word)
cv.imwrite('word_cut' + str(word_count) + '.jpg', word_cut)#保存分割出的二值化图像
cv.imwrite('cut_color' + str(word_count) + '.jpg', word_cut_color)#保存分割出的灰度图像
word_count+=1
word_start_a = word_end_a
print word_set
##### 合并与切分Clustering or Cutting #####
word_length_set = []
word_cut_position = []
#统计分割字段总长度、平均长度、最小长度
word_shortest_length = word_set[0][1] - word_set[0][0]
for j in range(0,len(word_set)):
word_length = word_set[j][1] - word_set[j][0]
word_total_length = word_total_length + word_length
word_length_set.append(word_length)
if word_length < word_shortest_length and float(word_b_set[row_cut][1] - word_b_set[row_cut][0])/word_length <= 2:
word_shortest_length = word_length
word_average_length = word_total_length/len(word_set)
#print word_set,len(word_set),cha_num
word_cluster_count = 0
if len(word_set) >= cha_num:#当分割个数超过设定的个数则偏向合并，否则切分
word_Cluster2Right = 0
for j in range(0,len(word_set)):
#当单个字段长度低于平均长度的0.75时，将字段向左/右合并，此处先标记合并后的新分割点
if word_length_set[j] < word_average_length*0.75:#1.25,0.75
if j != len(word_set)-1 and word_length_set[j-1] >= word_length_set[j+1]:#向右合并
if j-1 < 0:#考虑边缘情况
cut_position = word_set[j][0]
word_cut_position.append(cut_position)
word_Cluster2Right = 1
print 1
if j == len(word_set)-1 or word_length_set[j-1] < word_length_set[j+1]:#向左合并
if word_Cluster2Right == 1:#考虑前一块向右聚合的情况
cut_position = word_set[j][1]
word_cut_position.append(cut_position)
print 2
else:
if j-1 < 0:
cut_position = word_set[j][0]
word_cut_position.append(cut_position)
else:
if len(word_cut_position) == 0:
cut_position = word_set[j][1]
word_cut_position.append(cut_position)
else:
del word_cut_position[len(word_cut_position)-1]
cut_position = word_set[j][1]
word_cut_position.append(cut_position)
print 3
word_Cluster2Right = 0
else:
if j-1 < 0:#不合并，且在边缘位置
cut_position = word_set[j][0]
word_cut_position.append(cut_position)
cut_position = word_set[j][1]
word_cut_position.append(cut_position)
print 4
else:#不合并，不在边缘位置
cut_position = word_set[j][1]
word_cut_position.append(cut_position)
print 5
word_Cluster2Right = 0
#print word_cut_position
print word_cut_position
word_cluster_set = []
#word_cluster_count = 0
#根据合并后的新标记点对该行进行重新切分
for j in range(0,len(word_cut_position)-1):
word_cluster = (word_cut_position[j], word_cut_position[j+1], word_b_set[row_cut][0], word_b_set[row_cut][1])
word_cluster_set.append(word_cluster)
word_cluster_cut = ~th2[y + word_b_set[row_cut][0]:y + word_b_set[row_cut][1], x + word_cut_position[j]:x + word_cut_position[j+1]]
word_cluster_color_cut = shrinkTwoTimesTranslation_copy[y + word_b_set[row_cut][0]:y + word_b_set[row_cut][1], x + word_cut_position[j]:x + word_cut_position[j+1]]
cv.imwrite('cluster_word' + str(word_cluster_count) + '.jpg', word_cluster_cut)
word_cluster_cut_set.append(word_cluster_cut)
cv.imwrite('color_cluster' + str(word_cluster_count) + '.jpg', word_cluster_color_cut)
word_cluster_color_cut_set.append(word_cluster_color_cut)
word_cluster_count+=1
print word_cluster_set
else:
##### 切分 #####
word_over_TotalLength = 0
is_someword_over = 0
if len(word_set) == 1:#当为单字时
h = word_set[0][3] - word_set[0][2]
w = word_set[0][1] - word_set[0][0]
b = [0 for z in range(0, h)]
b_total = 0
single_cut = ~th2[y + word_set[0][2]:y + word_set[0][3], x + word_set[0][0]:x + word_set[0][1]]
#通过对整体水平、垂直投影量切去除无字空白部分
for j in range(0,h):
for i in range(0,w):
if single_cut[j,i] == 0:
b[j]+= 1
b_total+= 1
start_already = 0
b_before = 0
cut_finish = 0
for j in range(0,h):
b_before+= b[j]#统计该列前方的总水平投影量
b_after = b_total - b_before#统计该列后方的总水平投影量
if b_before > 4 and start_already == 0 and cut_finish != 1:
b_cut_start = j
start_already = 1
if start_already == 1 and b_after < 4 and cut_finish != 1:
b_cut_end = j
cut_finish = 1
a = [0 for z in range(0, w)]
a_total = 0
for j in range(0,w):
for i in range(0,h):
if single_cut[i,j] == 0:
a[j]+= 1
a_total+= 1
start_already = 0
a_before = 0
cut_finish = 0
for j in range(0,w):
a_before+= a[j]#统计该列前方的总垂直投影量
a_after = a_total - a_before#统计该列后方的总垂直投影量
if a_before > 4 and start_already == 0 and cut_finish != 1:
a_cut_start = j
start_already = 1
if start_already == 1 and a_after < 4 and cut_finish != 1:
a_cut_end = j
cut_finish = 1
#当垂直投影长度/水平投影长度大于2倍时，则切分为2个字段
print "len(word_set) == 1:",float(a_cut_end - a_cut_start)/(b_cut_end - b_cut_start)
if float(a_cut_end - a_cut_start)/(b_cut_end - b_cut_start) > 2:#1.8
word_part_length = int((a_cut_end - a_cut_start)/2)
for i in range(0,2):
word_cluster_cut = ~th2[y + b_cut_start + word_b_set[row_cut][0]:y + b_cut_end + word_b_set[row_cut][0], x + word_set[0][0] + a_cut_start + i*word_part_length:x + word_set[0][0] + a_cut_start + (i+1)*word_part_length]
word_cluster_color_cut = shrinkTwoTimesTranslation_copy[y + b_cut_start + word_b_set[row_cut][0]:y + b_cut_end + word_b_set[row_cut][0], x + word_set[0][0] + a_cut_start + i*word_part_length:x + word_set[0][0] + a_cut_start + (i+1)*word_part_length]
cv.imwrite('cluster_word' + str(word_cluster_count) + '.jpg', word_cluster_cut)
word_cluster_cut_set.append(word_cluster_cut)
cv.imwrite('color_cluster' + str(word_cluster_count) + '.jpg', word_cluster_color_cut)
word_cluster_color_cut_set.append(word_cluster_color_cut)
word_cluster_count+=1
else:#垂直投影长度/水平投影长度小于等于2倍时，则保持单字段
word_cluster_set = word_set
for j in range(0,len(word_cluster_set)):
word_cluster_cut = ~th2[y + word_b_set[row_cut][0]:y + word_b_set[row_cut][1], x + word_cluster_set[j][0]:x + word_cluster_set[j][1]]
word_cluster_color_cut = shrinkTwoTimesTranslation_copy[y + word_b_set[row_cut][0]:y + word_b_set[row_cut][1], x + word_cluster_set[j][0]:x + word_cluster_set[j][1]]
cv.imwrite('cluster_word' + str(word_cluster_count) + '.jpg', word_cluster_cut)
word_cluster_cut_set.append(word_cluster_cut)
cv.imwrite('color_cluster' + str(word_cluster_count) + '.jpg', word_cluster_color_cut)
word_cluster_color_cut_set.append(word_cluster_color_cut)
word_cluster_count+=1
else:#当为多个字时
for j in range(0,len(word_set)):#是否有超过最短字段长度2倍的字段
if word_length_set[j] > word_shortest_length*2.0 and len(word_set)>1:#1.75
word_over_TotalLength+= word_length_set[j]
is_someword_over = 1
#print word_over_TotalLength,word_average_length
if is_someword_over == 1:#若有超过最短字段2倍的字段，进入分割流程
#word_cluster_count = 0
word_cluster_set = word_set
for j in range(0,len(word_cluster_set)):
if word_length_set[j] > word_shortest_length*3.0 :#对长度超过最短字段长度3倍的字段进行平均切分
word_part_length = int(word_length_set[j]/math.ceil(float(word_length_set[j])/word_shortest_length))#word_average_length
for i in range(0,int(math.ceil(float(word_length_set[j])/word_shortest_length))):#word_average_length
word_cluster_cut = ~th2[y + word_b_set[row_cut][0]:y + word_b_set[row_cut][1], x + word_cluster_set[j][0] + i*word_part_length:x + word_cluster_set[j][0] + (i+1)*word_part_length]
word_cluster_color_cut = shrinkTwoTimesTranslation_copy[y + word_b_set[row_cut][0]:y + word_b_set[row_cut][1], x + word_cluster_set[j][0] + i*word_part_length:x + word_cluster_set[j][0] + (i+1)*word_part_length]
cv.imwrite('cluster_word' + str(word_cluster_count) + '.jpg', word_cluster_cut)
word_cluster_cut_set.append(word_cluster_cut)
cv.imwrite('color_cluster' + str(word_cluster_count) + '.jpg', word_cluster_color_cut)
word_cluster_color_cut_set.append(word_cluster_color_cut)
word_cluster_count+=1
else:#对长度不超过最短字段长度3倍的字段，保持原状
word_cluster_cut = ~th2[y + word_b_set[row_cut][0]:y + word_b_set[row_cut][1], x + word_cluster_set[j][0]:x + word_cluster_set[j][1]]
word_cluster_color_cut = shrinkTwoTimesTranslation_copy[y + word_b_set[row_cut][0]:y + word_b_set[row_cut][1], x + word_cluster_set[j][0]:x + word_cluster_set[j][1]]
cv.imwrite('cluster_word' + str(word_cluster_count) + '.jpg', word_cluster_cut)
word_cluster_cut_set.append(word_cluster_cut)
cv.imwrite('color_cluster' + str(word_cluster_count) + '.jpg', word_cluster_color_cut)
word_cluster_color_cut_set.append(word_cluster_color_cut)
word_cluster_count+=1
else:#若没有超过最短字段2倍的字段，则保持原状
word_cluster_set = word_set
for j in range(0,len(word_cluster_set)):
word_cluster_cut = ~th2[y + word_b_set[row_cut][0]:y + word_b_set[row_cut][1], x + word_cluster_set[j][0]:x + word_cluster_set[j][1]]
word_cluster_color_cut = shrinkTwoTimesTranslation_copy[y + word_b_set[row_cut][0]:y + word_b_set[row_cut][1], x + word_cluster_set[j][0]:x + word_cluster_set[j][1]]
cv.imwrite('cluster_word' + str(word_cluster_count) + '.jpg', word_cluster_cut)
word_cluster_cut_set.append(word_cluster_cut)
cv.imwrite('color_cluster' + str(word_cluster_count) + '.jpg', word_cluster_color_cut)
word_cluster_color_cut_set.append(word_cluster_color_cut)
word_cluster_count+=1
for j in range(0,w): #遍历每一列
for i in range((h-a[j]),h): #从该列应该变黑的最顶部的点开始向最底部涂黑
cut[i,j]=0 #涂黑
cv.imwrite('cut' + str(row_cut) + '.jpg', cut)
chinese_result = ''
last_cha = ''
last_cha_set = []
if is_ABO == 1:#若是在判断血型，因该Python文件用于识别籍贯部分的汉字，因此未用上该部分
### Last word detection ###
img = word_cluster_cut_set[len(word_cluster_cut_set)-1]
img_color = word_cluster_color_cut_set[len(word_cluster_cut_set)-1]
smaller = 0.9#0.9--normal,0.5--ABO
#Place to middle
h, w = img.shape
b = [0 for z in range(0, h)]
h_total = 0
h_num = 0
for j in range(0,h):
for i in range(0,w):
if img[j,i] == 0:
b[j]+= 1
h_num+= 1
h_total+= j
print "h_num:",h_num
if h_num != 0:
h_average = h_total//h_num
a = [0 for z in range(0, w)]
w_total = 0
w_num = 0
for j in range(0,w):
for i in range(0,h):
if img[i,j] == 0:
a[j]+= 1
w_num+= 1
w_total+= j
w_average = w_total//w_num
affineShrinkTranslation = np.array([[1, 0, int(w//2 - w_average)], [0, 1, int(h//2 - h_average)]], np.float32)
shrinkTwoTimesTranslation = cv.warpAffine(~img, affineShrinkTranslation, (w, h))
shrinkTwoTimesTranslation = cv.resize(shrinkTwoTimesTranslation,(int(w*smaller),int(h*smaller)),interpolation=cv.INTER_AREA)
cv.imwrite('bin_shrinkTwoTimesTranslation.jpg', shrinkTwoTimesTranslation)
shrinkTwoTimesTranslation_color = cv.warpAffine(~img_color, affineShrinkTranslation, (w, h))
shrinkTwoTimesTranslation_color = ~cv.resize(shrinkTwoTimesTranslation_color,(int(w*smaller),int(h*smaller)),interpolation=cv.INTER_AREA)
cv.imwrite('bin_shrinkTwoTimesTranslation_color.jpg', shrinkTwoTimesTranslation_color)
max_one = max(h,w)
bin = np.zeros((max_one,max_one), np.uint8)
bin.fill(0)
bin_color = np.zeros((max_one,max_one,3), np.uint8)
bin_color.fill(255)
rows_count = 0
for j in range(int(max_one//2-h*smaller/2), int(max_one//2+h*smaller/2)):
cols_count = 0
for i in range(int(max_one//2-w*smaller/2), int(max_one//2+w*smaller/2)):
if cols_count <= int(w*smaller)-1 and rows_count <= int(h*smaller)-1:
#print rows_count,cols_count,j,i
bin[j][i] = shrinkTwoTimesTranslation[rows_count][cols_count]
#print shrinkTwoTimesTranslation_color[rows_count][cols_count]
bin_color[j][i] = shrinkTwoTimesTranslation_color[rows_count][cols_count]
cols_count+= 1
rows_count+= 1
cv.imwrite('bin_ori.jpg', ~bin)
cv.imwrite('bin_ori_color.jpg', bin_color)
img = Image.fromarray(cv.cvtColor(bin_color,cv.COLOR_BGR2GRAY))
last_cha_set = chinese_ocr.inference(img, 1)
if len(last_cha_set) != 0:
#print len(last_cha_set)
if (last_cha_set[0] == 689 or last_cha_set[1] == 689 or last_cha_set[2] == 689):
last_cha = '型'
finding_len = len(word_cluster_cut_set) - 1
else:
last_cha = ''
finding_len = len(word_cluster_cut_set)
else:
last_cha = ''
finding_len = len(word_cluster_cut_set)
### Letter detection ###
for num in range(0,finding_len):
img = word_cluster_cut_set[num]
img_color = word_cluster_color_cut_set[num]
smaller = 0.5#0.9--normal,0.5--ABO
#Place to middle
h, w = img.shape
b = [0 for z in range(0, h)]
h_total = 0
h_num = 0
for j in range(0,h):
for i in range(0,w):
if img[j,i] == 0:
b[j]+= 1
h_num+= 1
h_total+= j
if h_num == 0:
continue
h_average = h_total//h_num
a = [0 for z in range(0, w)]
w_total = 0
w_num = 0
for j in range(0,w):
for i in range(0,h):
if img[i,j] == 0:
a[j]+= 1
w_num+= 1
w_total+= j
w_average = w_total//w_num
affineShrinkTranslation = np.array([[1, 0, int(w//2 - w_average)], [0, 1, int(h//2 - h_average)]], np.float32)
shrinkTwoTimesTranslation = cv.warpAffine(~img, affineShrinkTranslation, (w, h))
shrinkTwoTimesTranslation = cv.resize(shrinkTwoTimesTranslation,(int(w*smaller),int(h*smaller)),interpolation=cv.INTER_AREA)
cv.imwrite('bin_shrinkTwoTimesTranslation.jpg', shrinkTwoTimesTranslation)
shrinkTwoTimesTranslation_color = cv.warpAffine(~img_color, affineShrinkTranslation, (w, h))
shrinkTwoTimesTranslation_color = ~cv.resize(shrinkTwoTimesTranslation_color,(int(w*smaller),int(h*smaller)),interpolation=cv.INTER_AREA)
cv.imwrite('bin_shrinkTwoTimesTranslation_color.jpg', shrinkTwoTimesTranslation_color)
max_one = max(h,w)
bin = np.zeros((max_one,max_one), np.uint8)
bin.fill(0)
bin_color = np.zeros((max_one,max_one,3), np.uint8)
bin_color.fill(255)
rows_count = 0
for j in range(int(max_one//2-h*smaller/2), int(max_one//2+h*smaller/2)):
cols_count = 0
for i in range(int(max_one//2-w*smaller/2), int(max_one//2+w*smaller/2)):
if cols_count <= int(w*smaller)-1 and rows_count <= int(h*smaller)-1:
#print rows_count,cols_count,j,i
bin[j][i] = shrinkTwoTimesTranslation[rows_count][cols_count]
#print shrinkTwoTimesTranslation_color[rows_count][cols_count]
bin_color[j][i] = shrinkTwoTimesTranslation_color[rows_count][cols_count]
cols_count+= 1
rows_count+= 1
cv.imwrite('bin_ori.jpg', ~bin)
cv.imwrite('bin_ori_color.jpg', bin_color)
img = Image.fromarray(~bin)
#img = Image.fromarray(~bin)
ABO_result = ABO_mix.ABO_detection(img)
print ABO_result
chinese_result = chinese_result + str(ABO_result)
else:
#若不是在判断血型栏
for num in range(0,len(word_cluster_cut_set)):
img = word_cluster_cut_set[num]
img_color = word_cluster_color_cut_set[num]
smaller = 0.9#0.9--normal,0.5--ABO,0.8--giguan，单字缩小倍数
if smaller > 1:#若放大文字部分
h, w = img.shape
b = [0 for z in range(0, h)]
b_total = 0
for j in range(0,h):
for i in range(0,w):
if img[j,i] == 0:
b[j]+= 1
b_total+= 1
start_already = 0
b_before = 0
cut_finish = 0
for j in range(0,h):
b_before+= b[j]
b_after = b_total - b_before
if b_before > 4 and start_already == 0 and cut_finish != 1:
b_cut_start = j
start_already = 1
if start_already == 1 and b_after < 4 and cut_finish != 1:
b_cut_end = j
cut_finish = 1
a = [0 for z in range(0, w)]
a_total = 0
for j in range(0,w):
for i in range(0,h):
if img[i,j] == 0:
a[j]+= 1
a_total+= 1
start_already = 0
a_before = 0
cut_finish = 0
for j in range(0,w):
a_before+= a[j]
a_after = a_total - a_before
if a_before > 4 and start_already == 0 and cut_finish != 1:
a_cut_start = j
start_already = 1
if start_already == 1 and a_after < 4 and cut_finish != 1:
a_cut_end = j
cut_finish = 1
print a_cut_start,a_cut_end,b_cut_start,b_cut_end
#直接提取出矩形中有文字的部分，周围不留空白
roi_bigger = img[b_cut_start:b_cut_end,a_cut_start:a_cut_end]
cv.imwrite('roi_bigger.jpg', roi_bigger)
roi_bigger_color = img_color[b_cut_start:b_cut_end,a_cut_start:a_cut_end]
cv.imwrite('roi_bigger_color.jpg', roi_bigger_color)
bin = ~cv.resize(roi_bigger,(int((a_cut_end-a_cut_start)*smaller),int((b_cut_end-b_cut_start)*smaller)),interpolation=cv.INTER_AREA)
cv.imwrite('bin_shrinkTwoTimesTranslation.jpg', bin)
bin_color = cv.resize(roi_bigger_color,(int((a_cut_end-a_cut_start)*smaller),int((b_cut_end-b_cut_start)*smaller)),interpolation=cv.INTER_AREA)
cv.imwrite('bin_shrinkTwoTimesTranslation_color.jpg', bin_color)
else:#若缩小文字部分
#将文字部分平移到图像中央Place to middle
h, w = img.shape
b = [0 for z in range(0, h)]
h_total = 0
h_num = 0
for j in range(0,h):
for i in range(0,w):
if img[j,i] == 0:
b[j]+= 1
h_num+= 1
h_total+= j
if h_num == 0:
continue
h_average = h_total//h_num
a = [0 for z in range(0, w)]
w_total = 0
w_num = 0
for j in range(0,w):
for i in range(0,h):
if img[i,j] == 0:
a[j]+= 1
w_num+= 1
w_total+= j
w_average = w_total//w_num
#在二值化图像中，将文字部分平移到中央
affineShrinkTranslation = np.array([[1, 0, int(w//2 - w_average)], [0, 1, int(h//2 - h_average)]], np.float32)
shrinkTwoTimesTranslation = cv.warpAffine(~img, affineShrinkTranslation, (w, h))
shrinkTwoTimesTranslation = cv.resize(shrinkTwoTimesTranslation,(int(w*smaller),int(h*smaller)),interpolation=cv.INTER_AREA)
cv.imwrite('bin_shrinkTwoTimesTranslation.jpg', shrinkTwoTimesTranslation)
#在灰度图像中，将文字部分平移到中央
shrinkTwoTimesTranslation_color = cv.warpAffine(~img_color, affineShrinkTranslation, (w, h))
shrinkTwoTimesTranslation_color = ~cv.resize(shrinkTwoTimesTranslation_color,(int(w*smaller),int(h*smaller)),interpolation=cv.INTER_AREA)
cv.imwrite('bin_shrinkTwoTimesTranslation_color.jpg', shrinkTwoTimesTranslation_color)
max_one = max(h,w)
bin = np.zeros((max_one,max_one), np.uint8)
bin.fill(0)
bin_color = np.zeros((max_one,max_one,3), np.uint8)
bin_color.fill(255)
#缩小文字部分，并放置到一张更大的空白图像中
rows_count = 0
for j in range(int(max_one//2-h*smaller/2), int(max_one//2+h*smaller/2)):
cols_count = 0
for i in range(int(max_one//2-w*smaller/2), int(max_one//2+w*smaller/2)):
if cols_count <= int(w*smaller)-1 and rows_count <= int(h*smaller)-1:
#print rows_count,cols_count,j,i
bin[j][i] = shrinkTwoTimesTranslation[rows_count][cols_count]
#print shrinkTwoTimesTranslation_color[rows_count][cols_count]
bin_color[j][i] = shrinkTwoTimesTranslation_color[rows_count][cols_count]
cols_count+= 1
rows_count+= 1
'''
horizontalsize = cols / scale
horizontalStructure = cv.getStructuringElement(cv.MORPH_RECT, (horizontalsize, 1))
erosion = cv.erode(th2,horizontalStructure,iterations = 1)
dilation = cv.dilate(erosion,horizontalStructure,iterations = 1)
'''
cv.imwrite('bin_ori.jpg', ~bin)
cv.imwrite('bin_ori_color.jpg', bin_color)
img = Image.fromarray(cv.cvtColor(bin_color,cv.COLOR_BGR2GRAY))#OpenCV Mat转PIL
#img = Image.fromarray(~bin)
if get_all == 1:
chinese_set = chinese_ocr.inference(img, get_all)#手写汉字识别，对三个识别结果全接收
print "chinese_set", chinese_set
else:
if get_all == 2 and len(word_cluster_cut_set) == 1:
chinese_set = chinese_ocr.inference(img, 1)#分割字段唯一时民族识别，接收三个识别结果对应的字典序号
chinese = chinese_ocr.inference(img, 0)#分割字段唯一时，接收识别出的以汉字字符形式表示的最佳结果
return_change = str(chinese)
print "chinese_set", chinese_set
else:
chinese = chinese_ocr.inference(img, get_all)#手写汉字识别，只接收识别出的以汉字字符形式表示的最佳结果
if chinese_num_lim < 3:#控制识别出的汉字个数
chinese_result = chinese_result + str(chinese)
chinese_num_lim+=1
if get_all == 1 or return_change != 0:
return return_change, chinese_set
if get_all != 1:
if last_cha == '型':
chinese_result = chinese_result + '型'
return_change = 2
#print "Chinese_ocr:",chinese_result
return return_change, chinese_result