-
Notifications
You must be signed in to change notification settings - Fork 0
/
data_augmentation.py
57 lines (51 loc) · 1.48 KB
/
data_augmentation.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
from PIL import Image
import pandas as pd
import numpy as np
from keras.preprocessing.image import ImageDataGenerator
import glob
import matplotlib.pyplot as plt
import random
import imgaug.augmenters as iaa
def load_images_np(path):
dataset =[]
i = 0
for file in glob.glob(path):
dataset.append(np.array(Image.open(file).convert("L")))
i += 1
if i == 18:
break
dataset = np.array(dataset)
dataset = np.expand_dims(dataset, axis = 3)
return (dataset)
def plot_image_np(data):
fig, axes = plt.subplots(3, 6, sharex='col', sharey='row', squeeze=True)
plt.subplots_adjust(wspace=0.1, hspace=0.1)
plt.tight_layout(pad=0, w_pad=0, h_pad=0.0)
# rands = np.squeeze(data[np.random.choice(data.shape[0], 18)])
rands = np.squeeze(data)
for i, ax in enumerate(axes.flat):
img = Image.fromarray(rands[i])
ax.imshow(img)
ax.axis("off")
plt.show()
#
test = load_images_np("/Users/malluin/goinfre/testset/**/**")
# plot_image_np(test)
# data_gen = ImageDataGenerator()
seq = iaa.SomeOf((0,2), [iaa.GaussianBlur(sigma = (0,5)), iaa.AdditiveGaussianNoise(scale = (0,15))])
def imgaug(image):
return seq.augment_image(image)
train_gen = ImageDataGenerator(
vertical_flip = 1,
horizontal_flip = 1,
rotation_range = 90,
brightness_range = [0.3, 1.5],
shear_range = 10,
width_shift_range = .1,
height_shift_range = .1,
fill_mode ='constant',
preprocessing_function = imgaug)
train_gen.fit(test)
for batch in train_gen.flow(test):
print(batch.shape)
plot_image_np(batch)