-
Notifications
You must be signed in to change notification settings - Fork 0
/
data_augmentation.py
57 lines (47 loc) · 1.85 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 keras.preprocessing.image import ImageDataGenerator, img_to_array, load_img
import os
import shutil
def data_augmentation(input_dir, output_dir):
# List all classes (assuming each subdirectory represents a class)
classes = os.listdir(input_dir)
# Create the output directory if it doesn't exist
if not os.path.exists(output_dir):
os.makedirs(output_dir)
else:
# Clear existing files in the directory
shutil.rmtree(output_dir)
os.makedirs(output_dir)
# Dictionary to store file paths for each class
data = {cls: [] for cls in classes}
# Gather file paths for each class
for cls in classes:
cls_dir = os.path.join(input_dir, cls)
data[cls] = [os.path.join(cls_dir, file) for file in os.listdir(cls_dir)]
# ImageDataGenerator for augmentation
datagen = ImageDataGenerator(
rescale=1.0 / 255,
rotation_range=40,
width_shift_range=0.2,
height_shift_range=0.2,
shear_range=0.2,
zoom_range=0.2,
horizontal_flip=True,
fill_mode='nearest'
)
# Generate and save augmented images
for cls, files in data.items():
cls_augmented_dir = os.path.join(output_dir, cls)
os.makedirs(cls_augmented_dir)
for file in files:
img = load_img(file)
x = img_to_array(img)
x = x.reshape((1,) + x.shape)
i = 0
for batch in datagen.flow(x, batch_size=1, save_to_dir=cls_augmented_dir, save_prefix=cls, save_format='jpg'):
i += 1
if i >= 5: # Generate 5 augmented images for each original image
break
# Usage example:
input_directory = "D:/Drone/New folder (2)/asl_alphabet_train"
output_directory = "D:/Drone/New folder (2)/asl_alphabet_train_augmented"
data_augmentation(input_directory, output_directory)