-
Notifications
You must be signed in to change notification settings - Fork 21
/
Copy pathread_pixel_data.py
245 lines (224 loc) · 8.11 KB
/
read_pixel_data.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
import numpy as np
import matplotlib.pyplot as plt
import imageio
def get_row_col(label):
row_col=[]
for i in range(1,5):
row_col.append(np.where(label==i))
return row_col
def read_factor_data():
yinzi_name = ['altitude', 'aspect', 'fault', 'ndvi', 'plan', 'profile', 'rainfall',
'river', 'road', 'slope', 'spi', 'sti', 'twi', 'soil', 'landuse', 'lithology']
row = 2636
col = 2206
nums_factors = len(yinzi_name)
img = np.zeros((row, col, nums_factors))
i = 0
yinzi_type = []
for i in range(nums_factors):
# print(i)
name = yinzi_name[i]
PATH = 'C:/Users/75129/Desktop/mypy/demo_all/tif_yanshan/'
ttt = imageio.imread(PATH+name+'.tif')
# if yinzi_type_ori[i]=='str':
# plt.pyplot.figure()
# plt.pyplot.imshow(img[:,:,i])
ttt[ttt == -32768] = 0
# ttt[ttt==-3.402823e+038]=0
img[:, :, i] = ttt
# img_i=load_data[name]
# ma=np.max(img_i)
# mi=np.min(img_i)
# img[:,:,i]=(img_i-mi)/(ma-mi)
i = i+1
# pca = PCA(n_c
return img
def get_3d_x_save_memory(factors,window_size):
# windows_size must be Odd number
factors_3d=np.zeros([factors.shape[0],factors.shape[1],factors.shape[2]])
index=np.arange(0,window_size*window_size)
index=index.reshape([window_size,window_size])
# print(index)
loc_x=[]
loc_y=[]
r=(window_size-1)/2
for i in range(window_size*window_size):
loc=np.where(index==i)
x=loc[0]
y=loc[1]
x=r-x
y=r-y
loc_x.append(x)
loc_y.append(y)
# print(loc_x)
# print(loc_y)
for i in range(window_size*window_size):
print(i)
x_start=int(max(loc_x[i],0))
x_end=int(min(loc_x[i],-0))
y_start=int(max(loc_y[i],0))
y_end=int(min(loc_y[i],-0))
x_start_3d=-x_end
x_end_3d=-x_start
y_start_3d=-y_end
y_end_3d=-y_start
# print(x_start,x_end)
# print(y_start,y_end)
if x_end==0:
x_end=None
if y_end==0:
y_end=None
if x_end_3d==0:
x_end_3d=None
if y_end_3d==0:
y_end_3d=None
tmp=factors[x_start_3d:x_end_3d,y_start_3d:y_end_3d,:]
# print(tmp.shape)
# print(factors_3d[i,x_start_3d:x_end_3d,y_start_3d:y_end_3d,:].shape)
factors_3d[x_start:x_end,y_start:y_end,:]=tmp
np.save('D:/yanshan_new/pixel_traing_data/factors3d_'+str(window_size)+'_'+str(i)+'.npy',factors_3d)
# factors_3d=factors_3d.reshape([factors.shape[0],factors.shape[1],window_size,window_size,factors.shape[2]])
print(factors_3d.shape)
return None
def get_3d_x(factors,window_size):
# windows_size must be Odd number
factors_3d=np.zeros([window_size*window_size,factors.shape[0],factors.shape[1],factors.shape[2]])
index=np.arange(0,window_size*window_size)
index=index.reshape([window_size,window_size])
# print(index)
loc_x=[]
loc_y=[]
r=(window_size-1)/2
for i in range(window_size*window_size):
loc=np.where(index==i)
x=loc[0]
y=loc[1]
x=r-x
y=r-y
loc_x.append(x)
loc_y.append(y)
# print(loc_x)
# print(loc_y)
for i in range(window_size*window_size):
x_start=int(max(loc_x[i],0))
x_end=int(min(loc_x[i],-0))
y_start=int(max(loc_y[i],0))
y_end=int(min(loc_y[i],-0))
x_start_3d=-x_end
x_end_3d=-x_start
y_start_3d=-y_end
y_end_3d=-y_start
# print(x_start,x_end)
# print(y_start,y_end)
if x_end==0:
x_end=None
if y_end==0:
y_end=None
if x_end_3d==0:
x_end_3d=None
if y_end_3d==0:
y_end_3d=None
tmp=factors[x_start_3d:x_end_3d,y_start_3d:y_end_3d,:]
# print(tmp.shape)
# print(factors_3d[i,x_start_3d:x_end_3d,y_start_3d:y_end_3d,:].shape)
factors_3d[i,x_start:x_end,y_start:y_end,:]=tmp
factors_3d=factors_3d.reshape([window_size,window_size,factors.shape[0],factors.shape[1],factors.shape[2]])
factors_3d=np.swapaxes(factors_3d,0,2)
factors_3d=np.swapaxes(factors_3d,1,3)
print(factors_3d.shape)
return factors_3d
def get_tr_tt_data_3d(data_3d,tr_tt_img):
#1 tt_nolandslide 0
#2 tr_nolandslide 1
#3 tt_landslide 2
#4 tr_landslide 3
#data_3d.shape = (row,col,window_size,window_size,channel)
data_3d=data_3d.reshape([data_3d.shape[0]*data_3d.shape[1],data_3d.shape[2],data_3d.shape[3],data_3d.shape[4]])
tr_tt_img=tr_tt_img.reshape([tr_tt_img.shape[0]*tr_tt_img.shape[1],])
locs=get_row_col(tr_tt_img)
for i in range(4):
loc=locs[i]
# print(loc)
tmp_data=data_3d[loc,:,:,:]
tmp_data=np.squeeze(tmp_data)
print(tmp_data.shape)
if i==0:
tt_x=tmp_data
tt_y=np.zeros([tmp_data.shape[0],])
elif i==1:
tr_x=tmp_data
tr_y=np.zeros([tmp_data.shape[0],])
elif i==2:
tt_x=np.concatenate((tt_x,tmp_data),axis=0)
tt_y=np.concatenate((tt_y,np.ones([tmp_data.shape[0],])),axis=0)
elif i==3:
tr_x=np.concatenate((tr_x,tmp_data),axis=0)
tr_y=np.concatenate((tr_y,np.ones([tmp_data.shape[0],])),axis=0)
# print(tt_x.shape)
# print(tt_y)
return tr_x,tr_y,tt_x,tt_y
def get_tr_tt_data_3d_save_memory(data_3d_path,tr_tt_img,window_size):
#1 tt_nolandslide 0
#2 tr_nolandslide 1
#3 tt_landslide 2
#4 tr_landslide 3
#data_3d.shape = (row,col,channel)
train_x=[]
# train_y=[]
test_x=[]
# test_y=[]
tr_tt_img=tr_tt_img.reshape([tr_tt_img.shape[0]*tr_tt_img.shape[1],])
locs=get_row_col(tr_tt_img)
for i in range(window_size*window_size):
data_3d=np.load(data_3d_path+'factors3d_'+str(window_size)+'_'+str(i)+'.npy')
# data_3d=np.load('D:/yanshan_new/pixel_traing_data/factors3d_'+str(window_size)+'_'+str(i)+'.npy')
data_3d=data_3d.reshape([data_3d.shape[0]*data_3d.shape[1],data_3d.shape[2]])
for i in range(4):
loc=locs[i]
# print(loc)
tmp_data=data_3d[loc,:]
tmp_data=np.squeeze(tmp_data)
# print(tmp_data.shape)
if i==0:
tt_x=tmp_data
tt_y=np.zeros([tmp_data.shape[0],])
elif i==1:
tr_x=tmp_data
tr_y=np.zeros([tmp_data.shape[0],])
elif i==2:
tt_x=np.concatenate((tt_x,tmp_data),axis=0)
tt_y=np.concatenate((tt_y,np.ones([tmp_data.shape[0],])),axis=0)
elif i==3:
tr_x=np.concatenate((tr_x,tmp_data),axis=0)
tr_y=np.concatenate((tr_y,np.ones([tmp_data.shape[0],])),axis=0)
train_x.append(tr_x)
test_x.append(tt_x)
train_x=np.array(train_x)
test_x=np.array(test_x)
train_x=np.swapaxes(train_x,0,1)
test_x=np.swapaxes(test_x,0,1)
train_x=train_x.reshape([train_x.shape[0],window_size,window_size,train_x.shape[2]])
test_x=test_x.reshape([test_x.shape[0],window_size,window_size,test_x.shape[2]])
print(train_x.shape)
print(test_x.shape)
plt.imshow(train_x[10,:,:,0])
plt.show()
# print(tt_x.shape)
# print(tt_y)
return train_x,tr_y,test_x,tt_y
if __name__ == '__main__':
factors=read_factor_data()
window_size=7
print('begin get_3d_x')
factors_3d=get_3d_x(factors,window_size)
print('end get_3d_x')
label=imageio.imread('D:/yanshan_new/tr_and_tt.tif')
data_3d_path='D:/yanshan_new/pixel_traing_data/'
tr_x,tr_y,tt_x,tt_y=get_tr_tt_data_3d(factors_3d,label)
# tr_x,tr_y,tt_x,tt_y=get_tr_tt_data_3d(factors_3d,label)
# save memory
np.save('D:/yanshan_new/pixel_traing_data/factors_3d_'+str(window_size)+'.npy',factors_3d)
np.save('D:/yanshan_new/pixel_traing_data/tr_x_'+str(window_size)+'.npy',tr_x)
np.save('D:/yanshan_new/pixel_traing_data/tr_y_'+str(window_size)+'.npy',tr_y)
np.save('D:/yanshan_new/pixel_traing_data/tt_x_'+str(window_size)+'.npy',tt_x)
np.save('D:/yanshan_new/pixel_traing_data/tt_y_'+str(window_size)+'.npy',tt_y)