-
Notifications
You must be signed in to change notification settings - Fork 1
/
Copy pathutils.py
85 lines (78 loc) · 3.28 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
import tensorflow as tf
import numpy as np
import os
def make_mat(path1):
"""Read folder paths to a list"""
Data_folder = sorted(os.listdir(path1))
Data_path_mat = []
for f in Data_folder:
if not f.startswith('.'):
Data_path_mat.append(path1 + '/' + f)
return(Data_path_mat)
def select_views(x_train_images_path, row):
"""Select views belonging to given row."""
finalViews =[]
for view in x_train_images_path:
name = os.path.basename(view)
namelist = name.split('_')
if namelist[0] == str(row):
finalViews.append(view)
finalViews = sorted(finalViews, key=lambda f: int((f.split('_')[1]).split('.')[0]))
return(finalViews)
def read_image(filename):
"""Loads and returns a PNG image file with values in [0..1]."""
string = tf.io.read_file(filename)
image = tf.image.decode_image(string, channels=3)
image = tf.cast(image, tf.float32)
image /= 255
return(image)
def quantize_image(image):
"""Convert [0..1] float image to [0..255] uint8."""
image = tf.round(image * 255)
image = tf.saturate_cast(image, tf.uint8)
return image
def read_image_test(path1):
"""Loads and returns a PNG image file with values in [0..1]."""
img = read_image(path1)
[h,w,c]=np.shape(img)
if h!=512:
if h!=434:
if (w-528)%2==0:
img=img[:,6:-6,:]
else:
img=img[:,7:-6,:]
if (h-352)%2==0:
img=img[12:-12,:,:]
else:
img=img[12:-11,:,:]
else:
img=img[1:-1,1:,:]
return(img)
def save_img(path,org,img,row,col):
"""Saves an image to a PNG file."""
img = tf.squeeze(quantize_image(img))
if org:
filename=path+'/org_'+str(row)+'_'+str(col)+'.png'
else:
filename=path+'/rec_'+str(row)+'_'+str(col)+'.png'
string = tf.image.encode_png(img)
return tf.io.write_file(filename, string)
def preprocess_data(images_path):
"""Read data and create feature tensors."""
for i in range(8):
img = tf.expand_dims(read_image(images_path[i]),axis=0)
if i==0:
images = img
else:
images = tf.concat((images,img),axis=0)
row=tf.strings.to_number(tf.strings.split(tf.strings.split(images_path[0],'_' )[0],'/')[-1])
pos_x=tf.math.divide(row*tf.ones([64,64,3]),5)
batchfeature1 = tf.stack((images[0,:,:,:],images[5,:,:,:],pos_x,(1/5)*tf.ones([64,64,3])),axis=0)
batchfeature2 = tf.stack((images[1,:,:,:],images[5,:,:,:],pos_x,(2/5)*tf.ones([64,64,3])),axis=0)
batchfeature3 = tf.stack((images[2,:,:,:],images[5,:,:,:],pos_x,(3/5)*tf.ones([64,64,3])),axis=0)
batchfeature4 = tf.stack((images[3,:,:,:],images[5,:,:,:],pos_x,(4/5)*tf.ones([64,64,3])),axis=0)
batchfeature5 = tf.stack((images[4,:,:,:],images[5,:,:,:],pos_x,(5/5)*tf.ones([64,64,3])),axis=0)
batchfeature6 = tf.stack((images[5,:,:,:],images[5,:,:,:],pos_x,(6/5)*tf.ones([64,64,3])),axis=0)
batchfeature7 = tf.stack((images[6,:,:,:],images[5,:,:,:],pos_x,(7/5)*tf.ones([64,64,3])),axis=0)
batchfeature8 = tf.stack((images[7,:,:,:],images[5,:,:,:],pos_x,(8/5)*tf.ones([64,64,3])),axis=0)
return(images,batchfeature1,batchfeature2,batchfeature3,batchfeature4,batchfeature5,batchfeature6,batchfeature7,batchfeature8)