/
ocr_denoice_image.py
201 lines (156 loc) · 4.42 KB
/
ocr_denoice_image.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
#coding=utf8
import ImageFilter
import fcm
import kmeans
import ocr_segmentation
def MedianFilter(image,size=3):
'''
图像去噪
采用中值滤波
медианный фильтр median filter
'''
return image.filter(ImageFilter.MedianFilter(size))
def DenoiceImage(image):
'''
图像去噪
采用中值滤波
медианный фильтр median filter
'''
median_filter=ImageFilter.MedianFilter()
return median_filter.filter(image)
def Denoice_FCM_Normal(image):
"""
图像去噪
常规FCM
"""
img = image.copy()
pix = img.load()
width, height = img.size
dat = []
for x in range(width):
for y in range(height):
dat.append((pix[x, y], (x, y)))
f1 = fcm.FCM_normal(dat)
f1.Run()
print f1.counter
print f1.J()
for i in range(len(dat)):
x, y = dat[i][1]
if f1.u[1, i] > f1.u[0, i]:
pix[x, y] = 255
else:
pix[x, y] = 0
return img
def Denoice_FCM_Fast(image):
"""
图像去噪
常规FCM
"""
img = image.copy()
width, height = img.size
gray_hist = [0]*256
gray_hist_pix = [[] for i in range(256)]
pix = img.load()
for x in range(width):
for y in range(height):
gray_hist[pix[x, y]] += 1
gray_hist_pix[pix[x, y]].append((x, y))
#print pix[x, y], len(gray_hist_pix[pix[x, y]])
dat = []
for l in range(len(gray_hist)):
dat.append((l, gray_hist[l]))
f1 = fcm.FCM_fast(dat)
f1.Run()
print f1.counter
print f1.J()
print len(gray_hist_pix[127]), width*height
for i in range(len(dat)):
if f1.u[1, i] > f1.u[0, i]:
for pos in gray_hist_pix[i]:
pix[pos] = 255
else:
for pos in gray_hist_pix[i]:
pix[pos] = 0
return img
def Denoice_FCM_block(image):
img = image.copy()
pix = img.load()
f1 = fcm.FCM_image(img)
f1.Run()
width, height = img.size
for x in range(width):
for y in range(height):
if f1.u[0, x, y] > f1.u[1, x, y]:
pix[x, y] = 0
else:
pix[x, y] = 255
return img
def Denoise_kmeans(image):
img = image.copy()
width, height = img.size
pix = img.load()
block_list = []
data = []
checked = []
print "Get Block"
for x in range(width):
for y in range(height):
if pix[x, y] > 0:
if (x, y) in checked:
continue
block = ocr_segmentation.get_block(img, x, y)
checked += block
data.append(len(block))
block_list.append(block)
print "数据长度:", len(data)
k1 = kmeans.KMeans(3)
k1.SetCenter([1, 20, 100])
k1.SetData(data)
k1.Run()
for i in k1.Group[0]["data"]:
for x, y in block_list[i]:
pix[x, y] = 0
return img
def Denoise_kmeans_fast(image):
image_block = image.copy()
pix_image_block = image_block.load()
width, height = image_block.size
block_list = []
data = []
print "Get Block"
for x in range(width):
for y in range(height):
if pix_image_block[x, y] > 0:
#print "before:",(x, y),image_block.histogram()[255]
#print "id:",id(pix_image_block)
block = ocr_segmentation.get_block_fast(pix_image_block, x, y, width, height)
#print "after:", image_block.histogram()[255],len(block)
#sss = raw_input()
data.append(len(block))
block_list.append(block)
print "数据长度:", len(data)
k1 = kmeans.KMeans(3)
k1.SetCenter([1, 20, 100])
k1.SetData(data)
k1.Run()
img = image.copy()
pix = img.load()
for i in k1.Group[0]["data"]:
for x, y in block_list[i]:
pix[x, y] = 0
return img
def RemoveBlankBorder(image):
"""
删除整体图像边界处留白
"""
img = image.copy()
width, height = img.size
pix = img.load()
#处理左边界
for y in range(height):
ocr_segmentation.get_block_fast(pix, 0, y, width, height)
ocr_segmentation.get_block_fast(pix, width-1, y, width, height)
for x in range(width):
ocr_segmentation.get_block_fast(pix, x, 0, width, height)
ocr_segmentation.get_block_fast(pix, x, height-1, width, height)
return img