-
Notifications
You must be signed in to change notification settings - Fork 0
/
data_augmentation.py
52 lines (44 loc) · 1.52 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
from numpy import expand_dims
from keras.preprocessing.image import load_img
from keras.preprocessing.image import img_to_array
from keras.preprocessing.image import ImageDataGenerator
import matplotlib.pyplot as plt
from tqdm import tqdm
import cv2
import os
DATASET_PATH = 'full_dataset/'
datagen = ImageDataGenerator(
width_shift_range=0.2,
height_shift_range=0.2,
brightness_range=[0.3,1.7],
zoom_range=[0.8,1.2]
)
# # generate samples and plot
# for i in range(5):
# # define subplot
# plt.subplot(330 + 1 + i)
# # generate batch of images
# batch = it.next()
# # convert to unsigned integers for viewing
# image = batch[0].astype('uint8')
# # plot raw pixel data
# plt.imshow(image)
# # show the figure
# plt.show()
new_folder = 'augmented_dataset'
if not (os.path.isdir(new_folder)): os.mkdir(new_folder)
if not os.listdir(new_folder):
[os.mkdir(new_folder+'/'+class_name) for class_name in os.listdir(DATASET_PATH)]
print('generating 9 augmented images from each original image ...')
for class_name in tqdm(os.listdir(DATASET_PATH)):
image_number = 1
for image_name in os.listdir(DATASET_PATH + class_name):
img = load_img(DATASET_PATH + class_name + '/' + image_name)
data = img_to_array(img)
samples = expand_dims(data, 0)
it = datagen.flow(samples, batch_size=1)
for i in range(9):
batch = it.next()
image = batch[0].astype('uint8')
cv2.imwrite(new_folder+ '/'+class_name+'/'+class_name+'_'+str(image_number)+'_aug_'+str(i)+'.jpg', image)
image_number += 1