def apply_transform(self, x, transform_parameters):
"""Applies a transformation to an image according to given parameters.
# Arguments
x: 3D tensor, single image.
transform_parameters: Dictionary with string - parameter pairs
describing the transformation.
Currently, the following parameters
from the dictionary are used:
- `'theta'`: Float. Rotation angle in degrees.
- `'tx'`: Float. Shift in the x direction.
- `'ty'`: Float. Shift in the y direction.
- `'shear'`: Float. Shear angle in degrees.
- `'zx'`: Float. Zoom in the x direction.
- `'zy'`: Float. Zoom in the y direction.
- `'flip_horizontal'`: Boolean. Horizontal flip.
- `'flip_vertical'`: Boolean. Vertical flip.
- `'channel_shift_intencity'`: Float. Channel shift intensity.
- `'brightness'`: Float. Brightness shift intensity.
# Returns
A transformed version of the input (same shape).
"""
# x is a single image, so it doesn't have image number at index 0
img_row_axis = self.row_axis - 1
img_col_axis = self.col_axis - 1
img_channel_axis = self.channel_axis - 1
x = apply_affine_transform(x,
transform_parameters.get('theta', 0),
transform_parameters.get('tx', 0),
transform_parameters.get('ty', 0),
transform_parameters.get('shear', 0),
transform_parameters.get('zx', 1),
transform_parameters.get('zy', 1),
row_axis=img_row_axis,
col_axis=img_col_axis,
channel_axis=img_channel_axis,
fill_mode=self.fill_mode,
cval=self.cval,
order=self.interpolation_order)
if transform_parameters.get('channel_shift_intensity') is not None:
x = apply_channel_shift(
x, transform_parameters['channel_shift_intensity'],
img_channel_axis)
if transform_parameters.get('flip_horizontal', False):
x = flip_axis(x, img_col_axis)
if transform_parameters.get('flip_vertical', False):
x = flip_axis(x, img_row_axis)
if transform_parameters.get('brightness') is not None:
x = apply_brightness_shift(x, transform_parameters['brightness'])
return x
def test_apply_brightness_shift_error(monkeypatch):
monkeypatch.setattr(affine_transformations, 'ImageEnhance', None)
with pytest.raises(ImportError):
affine_transformations.apply_brightness_shift(0, [0])
def create_augment_videos() -> None:
images_paths = sorted(
imutils.paths.list_images(f'{HOME}/.keras/large-mpi-db')
)
for code in ['islf', 'kabf', 'lekf', 'milf', 'silf', 'cawm', 'chsm', 'jakm', 'juhm', 'mamm']:
for label in LABELS:
filtered_images_paths = list(
filter(
lambda image_path: image_path.split(os.path.sep)[-1].startswith(code)
and image_path.split(os.path.sep)[-2] == label,
images_paths
)
)
for t_count, transformation in enumerate(get_random_transformations()):
path = f'{HOME}/.keras/datasets/cec-videos-augmented/{label}'
video_path = f'{path}/{code}_{label}_{t_count:02d}.avi'
cap = cv2.VideoCapture(video_path)
if cap.get(cv2.CAP_PROP_FRAME_COUNT) == len(filtered_images_paths):
print(f'Skipping: {video_path}')
continue
os.makedirs(path, exist_ok=True)
out = cv2.VideoWriter(
video_path,
cv2.VideoWriter_fourcc(*'XVID'),
25,
(224, 224)
)
print(f'Creating: {video_path}')
for i_count, image_path in enumerate(filtered_images_paths):
x = cv2.imread(image_path)
x = cv2.resize(x, (224, 168))
x = cv2.copyMakeBorder(x, 28, 28, 0, 0, cv2.BORDER_CONSTANT)
if i_count == 0:
cv2.imwrite(f'{path}/transformation_oo.png', x)
x = apply_affine_transform(
x,
transformation.get('theta', 0),
transformation.get('tx', 0),
transformation.get('ty', 0),
transformation.get('shear', 0),
transformation.get('zx', 1),
transformation.get('zy', 1)
)
if transformation.get('channel_shift_intensity') is not None:
x = apply_channel_shift(
x,
transformation['channel_shift_intensity']
)
if transformation.get('flip_horizontal', False):
x = flip_axis(x, 1)
if transformation.get('flip_vertical', False):
x = flip_axis(x, 0)
if transformation.get('brightness') is not None:
x = apply_brightness_shift(x, transformation['brightness'])
x = np.uint8(x)
if transformation.get('grayscale', False):
x = cv2.cvtColor(x, cv2.COLOR_RGB2GRAY)
x = np.repeat(x[:, :, np.newaxis], 3, axis=2)
if i_count == 0:
cv2.imwrite(f'{path}/transformation_{t_count:02d}.png', x)
out.write(x)
out.release()
return