def test_augment_images__deterministic(self):
rs = np.random.RandomState(1)
images = rs.randint(0, 255, size=(32, 4, 4, 3), dtype=np.uint8)
for deterministic in [False, True]:
aug = iaa.MultiplyHueAndSaturation(mul=(0.1, 5.0),
deterministic=deterministic)
images_aug1 = aug.augment_images(images)
images_aug2 = aug.augment_images(images)
equal = np.array_equal(images_aug1, images_aug2)
if deterministic:
assert equal
else:
assert not equal
aug = iaa.MultiplyHueAndSaturation(mul_hue=(0.1, 5.0),
mul_saturation=(0.1, 5.0),
deterministic=deterministic)
images_aug1 = aug.augment_images(images)
images_aug2 = aug.augment_images(images)
equal = np.array_equal(images_aug1, images_aug2)
if deterministic:
assert equal
else:
assert not equal
def main():
image = ia.quokka_square((128, 128))
images_aug = []
for mul in np.linspace(0.0, 2.0, 10):
aug = iaa.MultiplyHueAndSaturation(mul)
image_aug = aug.augment_image(image)
images_aug.append(image_aug)
for mul_hue in np.linspace(0.0, 5.0, 10):
aug = iaa.MultiplyHueAndSaturation(mul_hue=mul_hue)
image_aug = aug.augment_image(image)
images_aug.append(image_aug)
for mul_saturation in np.linspace(0.0, 5.0, 10):
aug = iaa.MultiplyHueAndSaturation(mul_saturation=mul_saturation)
image_aug = aug.augment_image(image)
images_aug.append(image_aug)
ia.imshow(ia.draw_grid(images_aug, rows=3))
images_aug = []
images_aug.extend(iaa.MultiplyHue().augment_images([image] * 10))
images_aug.extend(iaa.MultiplySaturation().augment_images([image] * 10))
ia.imshow(ia.draw_grid(images_aug, rows=2))
def __init__(self,
coco_dir,
set_name='val2014',
img_size=416,
multiscale=False,
phase='Test'):
super().__init__(coco_dir, set_name, img_size)
self.coco_dir = coco_dir
self.set_name = set_name
self.img_size = img_size
self.max_objects = 100
self.min_size = img_size - 3 * 32
self.max_size = img_size + 3 * 32
self.multiscale = multiscale
self.batch_count = 0
self.augmentation = False
if phase == 'Train':
self.augmentation = True
self.aug_seq = iaa.Sequential([
iaa.Affine(
scale=(0.8, 1.2),
rotate=(-10, 10),
shear=(-15, 15),
translate_percent=(-0.05, 0.05),
),
iaa.Fliplr(p=0.5),
iaa.MultiplyHueAndSaturation((0.5, 1.5), per_channel=True),
])
self.mosaic = self.augmentation
def __init__(self):
self.seq = iaa.Sequential(
[
# iaa.Fliplr(0.5), # horizontal flips
# Small gaussian blur with random sigma between 0 and 0.5.
# But we only blur about 50% of all images.
iaa.GaussianBlur(sigma=(0, 0.5)),
iaa.MotionBlur(k=[5, 12], angle=[-45, 45]),
# Strengthen or weaken the contrast in each image.
iaa.Alpha([0.25, 0.35, 0.55],
iaa.Sequential([
iaa.GaussianBlur(sigma=(60, 100)),
iaa.LinearContrast((1, 3)),
iaa.Add((0, 30))
])),
#iaa.Lambda(radial_blur),
# Add gaussian noise.
# For 50% of all images, we sample the noise once per pixel.
# For the other 50% of all images, we sample the noise per pixel AND
# channel. This can change the color (not only brightness) of the
# pixels.
iaa.LinearContrast((0.5, 1.0)),
iaa.MultiplyHueAndSaturation((0.5, 1.5))
# iaa.Alpha([0.25, 0.35], iaa.Clouds()),
],
random_order=False)
def __init__(self, data_root, image_shape, plus1, mode, \
use_subset=False, no_flow=False, precomputed_val=None, \
sample_length=5):
self.no_flow = no_flow
self.mode = mode
self.precomputed_val = precomputed_val
self.sample_length = sample_length
assert self.mode in ['train', 'val']
if self.precomputed_val is not None:
assert self.mode == 'val'
self.data_root = data_root
if plus1:
self.image_shape = [image_shape[0] + 1, image_shape[1] + 1]
else:
self.image_shape = list(image_shape)
setname = '{}_videos_subset.txt' if use_subset else '{}_videos.txt'
setname = setname.format(self.mode)
with open(os.path.join(self.data_root, 'frame_corr.json'), 'r') as f:
self.frame_corr = json.load(f)
with open(os.path.join(self.data_root, setname), 'r') as f:
self.samples = self.parse(f)
#self.samples = self.samples[:240]
self.dataset_length = len(self.samples)
# apply to fg, bg
self.pixel_aug = iaa.Sequential([
iaa.MultiplyHueAndSaturation(mul=iap.TruncatedNormal(
1.0, 0.2, 0.5, 1.5)), # mean, std, low, high
iaa.GammaContrast(gamma=iap.TruncatedNormal(1.0, 0.2, 0.5, 1.5)),
iaa.AddToHue(value=iap.TruncatedNormal(0.0, 0.1 * 100, -0.2 *
255, 0.2 * 255)),
])
self.jpeg_aug = iaa.Sometimes(
0.6, iaa.JpegCompression(compression=(70, 99)))
def YOLO():
"""
Data augmentation model for YOLOv3 training
"""
return iaa.Sequential([
iaa.KeepSizeByResize(
iaa.Affine(
scale=iap.Normal(1, 0.125),
translate_percent=0.1,
cval=128,
)),
iaa.Fliplr(0.5),
iaa.Resize({
"height": iap.Normal(1, 0.1),
"width": iap.Normal(1, 0.1)
}),
iaa.Resize({
"longer-side": 416,
"shorter-side": "keep-aspect-ratio"
}),
iaa.PadToFixedSize(416, 416, pad_cval=128),
iaa.MultiplyHueAndSaturation(mul_hue=iap.Uniform(0, 2),
mul_saturation=iap.Uniform(1 / 1.5, 1.5)),
iaa.AssertShape((None, 416, 416, 3)),
])
def __init__(self, data_dir, batch_size, image_shape, training=True):
"""
Keras Sequence object to train a model on larger-than-memory data.
@:param: data_dir: directory in which we have got the kitti images and the corresponding masks
@:param: batch_size: define the number of training samples to be propagated.
@:param: image_shape: shape of the input image
"""
self.batch_size = batch_size
self.image_shape = image_shape
self.training = training
self.image_paths = glob(os.path.join(data_dir, 'image_2', '*.png'))
#print(self.image_paths)
print("*****************[DATA INFO]*****************")
if (training):
print("Found " + str(len(self.image_paths)) + " training images")
else:
print("Found " + str(len(self.image_paths)) + " validation images")
print("*********************************************")
if (training):
self.label_paths = {re.sub(r'_(lane|road)_', '_', os.path.basename(path)): path
for path in glob(os.path.join(data_dir, 'gt_image_2', '*_road_*.png'))}
#glob(os.path.join(data_dir, 'gt_image_2', '*.png'))
else:
self.label_paths = {os.path.basename(path): path
for path in glob(os.path.join(data_dir, 'gt_image_2', '*.png'))}
#print(self.label_paths)
self.sometimes = lambda aug: iaa.Sometimes(0.5, aug)
self.aug_pipe = iaa.Sequential(
[
iaa.SomeOf((0, 5),
[
iaa.OneOf([
iaa.GaussianBlur((0, 3.0)), # blur images with a sigma between 0 and 3.0
iaa.AverageBlur(k=(2, 7)),
# blur image using local means with kernel sizes between 2 and 7
iaa.MedianBlur(k=(3, 11)),
# blur image using local medians with kernel sizes between 2 and 7
]),
iaa.Sharpen(alpha=(0, 1.0), lightness=(0.75, 1.5)), # sharpen images
iaa.AdditiveGaussianNoise(loc=0, scale=(0.0, 0.05 * 255), per_channel=0.5),
iaa.OneOf([
iaa.Dropout((0.01, 0.1), per_channel=0.5), # randomly remove up to 10% of the pixels
]),
iaa.Add((-10, 10), per_channel=0.5),
# change brightness of images (by -10 to 10 of original value)
iaa.Multiply((0.5, 1.5), per_channel=0.5),
iaa.ContrastNormalization((0.5, 2.0), per_channel=0.5), # improve or worsen the contrast
iaa.MultiplyHueAndSaturation((0.5, 1.5)),
],
random_order=True
)
],
random_order=True
)
def hue_transform1(self, image):
z = image.shape[0]
para = random.randrange(6, 14, 1)
hue_m = para / 10
aug = iaa.MultiplyHueAndSaturation((hue_m, hue_m))
for stack in range(z):
image[stack] = aug.augment_image(image[stack])
return image
def chapter_augmenters_multiplyhueandsaturation():
fn_start = "color/multiplyhueandsaturation"
aug = iaa.MultiplyHueAndSaturation((0.5, 1.5), per_channel=True)
run_and_save_augseq(fn_start + ".jpg",
aug, [ia.quokka(size=(128, 128)) for _ in range(8)],
cols=4,
rows=2)
aug = iaa.MultiplyHueAndSaturation(mul_hue=(0.5, 1.5))
run_and_save_augseq(fn_start + "_mul_hue.jpg",
aug, [ia.quokka(size=(128, 128)) for _ in range(8)],
cols=4,
rows=2)
aug = iaa.MultiplyHueAndSaturation(mul_saturation=(0.5, 1.5))
run_and_save_augseq(fn_start + "_mul_saturation.jpg",
aug, [ia.quokka(size=(128, 128)) for _ in range(8)],
cols=4,
rows=2)
def test_returns_correct_objects__mul(self):
aug = iaa.MultiplyHueAndSaturation((0.9, 1.1), per_channel=True)
assert isinstance(aug, iaa.WithHueAndSaturation)
assert isinstance(aug.children, iaa.Sequential)
assert len(aug.children) == 1
assert isinstance(aug.children[0], iaa.Multiply)
assert isinstance(aug.children[0].mul, iap.Uniform)
assert np.isclose(aug.children[0].mul.a.value, 0.9)
assert np.isclose(aug.children[0].mul.b.value, 1.1)
assert isinstance(aug.children[0].per_channel, iap.Deterministic)
assert aug.children[0].per_channel.value == 1
def test_returns_correct_objects__mul_saturation(self):
aug = iaa.MultiplyHueAndSaturation(mul_saturation=(0.9, 1.1))
assert isinstance(aug, iaa.WithHueAndSaturation)
assert isinstance(aug.children, iaa.Sequential)
assert len(aug.children) == 1
assert isinstance(aug.children[0], iaa.WithChannels)
assert aug.children[0].channels == [1]
assert len(aug.children[0].children) == 1
assert isinstance(aug.children[0].children[0], iaa.Multiply)
assert isinstance(aug.children[0].children[0].mul, iap.Uniform)
assert np.isclose(aug.children[0].children[0].mul.a.value, 0.9)
assert np.isclose(aug.children[0].children[0].mul.b.value, 1.1)
def ol_aug(image, mask):
# ia.seed(seed)
# Example batch of images.
# The array has shape (32, 64, 64, 3) and dtype uint8.
images = image # B,H,W,C
masks = mask # B,H,W,C
# print('In Aug',images.shape,masks.shape)
combo = np.concatenate((images, masks), axis=3)
# print('COMBO: ',combo.shape)
seq_all = iaa.Sequential([
iaa.Fliplr(0.5), # horizontal flips
# iaa.PadToFixedSize(width=crop_size[0], height=crop_size[1]),
# iaa.CropToFixedSize(width=crop_size[0], height=crop_size[1]),
iaa.Affine(
scale={"x": (0.9, 1.1), "y": (0.9, 1.1)},
# scale images to 90-110% of their size, individually per axis
translate_percent={"x": (-0.1, 0.1), "y": (-0.1, 0.1)},
# translate by -10 to +10 percent (per axis)
rotate=(-5, 5), # rotate by -5 to +5 degrees
shear=(-3, 3), # shear by -3 to +3 degrees
),
# iaa.Cutout(nb_iterations=(1, 5), size=0.2, cval=0, squared=False),
], random_order=False) # apply augmenters in random order
seq_f = iaa.Sequential([
iaa.Sometimes(0.5,
iaa.OneOf([
iaa.GaussianBlur((0.0, 3.0)),
iaa.MotionBlur(k=(3, 20)),
]),
),
iaa.Sometimes(0.5,
iaa.OneOf([
iaa.Multiply((0.8, 1.2), per_channel=0.2),
iaa.MultiplyBrightness((0.5, 1.5)),
iaa.LinearContrast((0.5, 2.0), per_channel=0.2),
iaa.BlendAlpha((0., 1.), iaa.HistogramEqualization()),
iaa.MultiplyHueAndSaturation((0.5, 1.5), per_channel=0.2),
]),
),
], random_order=False)
combo_aug = np.array(seq_all.augment_images(images=combo))
# print('combo_au: ', combo_aug.shape)
images_aug = combo_aug[:, :, :, :3]
masks_aug = combo_aug[:, :, :, 3:]
images_aug = seq_f.augment_images(images=images_aug)
return images_aug, masks_aug
def data_augmentation(path):
ia.seed(2)
seq = iaa.Sequential([
iaa.Sometimes(0.5, iaa.Grayscale(alpha=(0.1, 0.5))),
iaa.Sometimes(0.5, iaa.Multiply((0.5, 1.5), per_channel=0.5)),
iaa.Sometimes(0.5,
iaa.MultiplyHueAndSaturation(mul_saturation=(0.5, 1.5))),
iaa.Sometimes(0.5, iaa.GaussianBlur(sigma=(0, 2.0))),
iaa.Sometimes(0.8, iaa.MultiplyBrightness((0.5, 1.5))),
iaa.AddToBrightness((-30, 30)),
iaa.Sometimes(0.6,
iaa.MultiplyHueAndSaturation(mul_saturation=(0.5, 1.5)))
],
random_order=True)
i = 0
for fname in os.listdir(path):
try:
img = imageio.imread(os.path.join(path, fname), pilmode="RGB")
print(i)
if i % 5 == 0:
img_aug = seq.augment_image(img)
imageio.imwrite(
os.path.join(path, fname.replace(".jpg", "_imgaug.jpg")),
img_aug)
fname_txt = fname.replace('.jpg', '.txt')
print(
os.path.join(path,
fname_txt.replace(".txt", "_imgaug.txt")))
shutil.copyfile(
os.path.join(path, fname_txt),
os.path.join(path,
fname_txt.replace(".txt", "_imgaug.txt")))
except:
print('Error reading img')
i += 1
def augmentor(images):
'Apply data augmentation'
seq = iaa.Sequential(
[
# apply the following augmenters to most images
iaa.Fliplr(0.9), # horizontally flip 50% of all images
iaa.Flipud(0.9), # vertically flip 20% of all images
iaa.ElasticTransformation(alpha=(10, 20), sigma=6),
iaa.GammaContrast((0.8, 1.2)),
iaa.MultiplyHueAndSaturation(mul_hue=(0.6, 1.4)),
iaa.GaussianBlur((0.0, 0.8)),
],
random_order=True)
return seq.augment_images(images)
def __init__(self, key_source='image', key_target=None):
super(RandomColorJitter, self).__init__(key_source=key_source,
key_target=key_target)
self.sequence = iaa.Sequential([
iaa.Sometimes(
0.8,
iaa.Sequential([
iaa.MultiplyBrightness((0.8, 1.25)),
iaa.MultiplyHueAndSaturation(mul_hue=(0.8, 1.25),
mul_saturation=(0.8, 1.25))
])),
iaa.Sometimes(0.2, iaa.Grayscale())
])
def test_augment_images__mul_hue_and_mul_saturation(self):
# this is almost identical to test_augment_images__mul
# only
# aug = ...
# and
# image_hsv[...] *= 1.2
# have been changed
aug = iaa.MultiplyHueAndSaturation(mul_hue=1.5,
mul_saturation=1.6) # changed
# example image
image = np.arange(0, 255).reshape((1, 255, 1)).astype(np.uint8)
image = np.tile(image, (1, 1, 3))
image[..., 0] += 0
image[..., 1] += 5
image[..., 2] += 10
# compute expected output
image_hsv = cv2.cvtColor(image, cv2.COLOR_RGB2HSV)
image_hsv = image_hsv.astype(np.int16) # simulate WithHueAndSaturation
image_hsv[..., 0] = ((image_hsv[..., 0].astype(np.float32) / 180) *
255).astype(np.int16)
image_hsv = image_hsv.astype(np.float32) # simulate Multiply
image_hsv[..., 0] *= 1.5
image_hsv[..., 1] *= 1.6 # changed over __mul
image_hsv = np.round(image_hsv).astype(np.int16)
image_hsv[..., 0] = np.mod(image_hsv[..., 0], 255)
image_hsv[..., 0] = ((image_hsv[..., 0].astype(np.float32) / 255) *
180).astype(np.int16)
image_hsv[..., 1] = np.clip(image_hsv[..., 1], 0, 255)
image_hsv = image_hsv.astype(np.uint8)
image_expected = cv2.cvtColor(image_hsv, cv2.COLOR_HSV2RGB)
assert not np.array_equal(image_expected, image)
# augment and verify
images_aug = aug.augment_images(np.stack([image, image], axis=0))
assert ia.is_np_array(images_aug)
for image_aug in images_aug:
assert image_aug.shape == (1, 255, 3)
diff = np.abs(image_aug.astype(np.int16) - image_expected)
assert np.all(diff <= 1)
def _load_augmentation_aug_non_geometric():
return iaa.Sequential([
iaa.Sometimes(0.3, iaa.Multiply((0.5, 1.5), per_channel=0.5)),
iaa.Sometimes(0.2, iaa.JpegCompression(compression=(70, 99))),
iaa.Sometimes(0.2, iaa.GaussianBlur(sigma=(0, 3.0))),
iaa.Sometimes(0.2, iaa.MotionBlur(k=15, angle=[-45, 45])),
iaa.Sometimes(0.2, iaa.MultiplyHue((0.5, 1.5))),
iaa.Sometimes(0.2, iaa.MultiplySaturation((0.5, 1.5))),
iaa.Sometimes(
0.34, iaa.MultiplyHueAndSaturation((0.5, 1.5), per_channel=True)),
iaa.Sometimes(0.34, iaa.Grayscale(alpha=(0.0, 1.0))),
iaa.Sometimes(0.2, iaa.ChangeColorTemperature((1100, 10000))),
iaa.Sometimes(0.1, iaa.GammaContrast((0.5, 2.0))),
iaa.Sometimes(0.2, iaa.SigmoidContrast(gain=(3, 10),
cutoff=(0.4, 0.6))),
iaa.Sometimes(0.1, iaa.CLAHE()),
iaa.Sometimes(0.1, iaa.HistogramEqualization()),
iaa.Sometimes(0.2, iaa.LinearContrast((0.5, 2.0), per_channel=0.5)),
iaa.Sometimes(0.1, iaa.Emboss(alpha=(0, 1.0), strength=(0, 2.0)))
])
def load_augmentation_aug_non_geometric():
return iaa.Sequential([
iaa.Sometimes(
0.5,
iaa.AdditiveGaussianNoise(loc=0,
scale=(0.0, 0.05 * 255),
per_channel=0.5)),
iaa.Sometimes(
0.5,
iaa.OneOf([
iaa.GaussianBlur(sigma=(0.0, 3.0)),
iaa.GaussianBlur(sigma=(0.0, 5.0))
])),
iaa.Sometimes(0.5, iaa.MultiplyAndAddToBrightness(mul=(0.4, 1.7))),
iaa.Sometimes(0.5, iaa.GammaContrast((0.4, 1.7))),
iaa.Sometimes(0.5, iaa.Multiply((0.4, 1.7), per_channel=0.5)),
iaa.Sometimes(0.5, iaa.MultiplyHue((0.4, 1.7))),
iaa.Sometimes(
0.5, iaa.MultiplyHueAndSaturation((0.4, 1.7),
per_channel=True)),
iaa.Sometimes(0.5, iaa.LinearContrast((0.4, 1.7), per_channel=0.5))
])
def __init__(self, da):
assert check_data_aug(da), 'Parameters for custom data augmentation missing. Should have: {}'.format(
keys_data_aug)
aug_geometric = [iaa.Affine(scale=(da['scalem'], da['scaleM']),
translate_percent={'x': (-da['trans'], da['trans']),
'y': (-da['trans'], da['trans'])},
rotate=(-da['rot'], da['rot']),
shear=(-da['shear'], da['shear']),
cval=FILL_COLOR),
iaa.PerspectiveTransform(scale=(0, da['pers']),
cval=FILL_COLOR, keep_size=True)]
aug_camera = [iaa.GaussianBlur(sigma=(0, da['sigma'])),
iaa.MotionBlur(k=(da['mot_km'], da['mot_kM']),
angle=(-da['mot_an'], da['mot_an']),
direction=(da['mot_dm'], da['mot_dM'])),
iaa.JpegCompression(compression=(da['jpegm'], da['jpegM'])),
iaa.LinearContrast(alpha=(da['con_alpham'], da['con_alphaM']),
per_channel=da['con_chan']),
iaa.MultiplyHueAndSaturation(mul=(da['col_mulm'], da['col_mulM']),
per_channel=da['col_chan'],
from_colorspace='BGR'),
iaa.AddToHueAndSaturation((da['col_addm'], da['col_addM']),
per_channel=da['col_chan'])
]
cutout = iaa.Cutout(nb_iterations=(0, da['co_num']),
size=(da['co_sm'], da['co_sM']),
squared=False,
cval=FILL_COLOR)
# Create a mix of all others
self.augmenter = iaa.Sequential([iaa.SomeOf((0, 1), aug_geometric), # none or 1
iaa.SomeOf((0, len(aug_camera) - 2), aug_camera), # from none to all-2
cutout], random_order=True) # mix the apply order
.loc[lambda df: df.label == "ASSIETTE_26"]
.pipe(RandomAssignment("tray_name"))
)
train_set = all_annotations.loc[lambda df: df.random_split == "train"]
support_set = train_set.assign(crop_coordinates=lambda df: df[["x1", "y1", "x2", "y2"]].agg(list, axis=1))
val_set = all_annotations.loc[lambda df: df.random_split == "val"]
#%% Init training
query_preprocessing = iaa.Sequential(
[
iaa.Resize({"longer-side": 416, "shorter-side": "keep-aspect-ratio"}),
iaa.PadToFixedSize(416, 416),
iaa.Fliplr(0.5),
iaa.Flipud(0.5),
iaa.MultiplyHueAndSaturation(mul_hue=iap.Uniform(0, 2), mul_saturation=iap.Uniform(1 / 1.5, 1.5)),
iaa.AssertShape((None, 416, 416, 3)),
]
)
support_preprocessing = iaa.Sequential(
[
iaa.Fliplr(0.5),
iaa.Flipud(0.5),
iaa.Affine(rotate=(-180, 180)),
iaa.Resize({"longer-side": 128, "shorter-side": "keep-aspect-ratio"}),
iaa.PadToFixedSize(128, 128, pad_mode="symmetric"),
iaa.MultiplyHueAndSaturation(mul_hue=iap.Uniform(0, 2), mul_saturation=iap.Uniform(1 / 1.5, 1.5)),
iaa.AssertShape((None, 128, 128, 3)),
]
)
##augmentation functions used in sequences pipeline
seq1 = iaa.Sequential([
iaa.AdditiveGaussianNoise(scale=(0, 0.3), per_channel=True),
iaa.Add((-25, 60)),
iaa.JpegCompression(compression=(30, 87)),
iaa.MedianBlur(k=(1, 3)),
iaa.PiecewiseAffine(scale=(0.01, 0.05)),
iaa.Fliplr(0.5),
iaa.Flipud(0.5),
iaa.CoarseDropout((0.0, 0.09), size_percent=(0.02, 0.15))
],
random_order=True)
seq2 = iaa.Sequential([
iaa.GaussianBlur(sigma=(0, 3)),
iaa.MultiplyHueAndSaturation((0.5, 1.5)),
iaa.JpegCompression(compression=(45, 87)),
iaa.PiecewiseAffine(scale=(0.01, 0.05)),
iaa.SaltAndPepper(p=(0.1, 0.15), per_channel=True),
iaa.Fliplr(0.5),
iaa.Flipud(0.5),
iaa.CoarseDropout((0.0, 0.09), size_percent=(0.02, 0.15))
],
random_order=True)
##divide the dataset into two batches
first_image_list = []
second_image_list = []
first_segmap_list = []
second_segmap_list = []
j = len(image_name_list)
path = 'idcard/'
sqe_list = [
iaa.ChangeColorspace(from_colorspace="RGB", to_colorspace="HSV"),
iaa.WithChannels(0, iaa.Add((-50, 50))),
iaa.WithChannels(1, iaa.Add((-50, 50))),
iaa.WithChannels(2, iaa.Add((-50, 50))),
iaa.ChangeColorspace(from_colorspace="HSV", to_colorspace="RGB"),
iaa.Add((-80, 80), per_channel=0.5),
iaa.Multiply((0.5, 1.5), per_channel=0.5),
iaa.AverageBlur(k=((5), (1, 3))),
iaa.AveragePooling(2),
iaa.AddElementwise((-20, -5)),
iaa.AdditiveGaussianNoise(scale=(0, 0.05 * 255)),
iaa.JpegCompression(compression=(50, 99)),
iaa.MultiplyHueAndSaturation(mul_hue=(0.5, 1.5)),
iaa.WithBrightnessChannels(iaa.Add((-50, 50))),
iaa.WithBrightnessChannels(iaa.Add((-50, 50)),
to_colorspace=[iaa.CSPACE_Lab, iaa.CSPACE_HSV]),
iaa.MaxPooling(2),
iaa.MinPooling((1, 2)),
# iaa.Superpixels(p_replace=(0.1, 0.2), n_segments=(16, 128)),
iaa.Clouds(),
iaa.Fog(),
iaa.AdditiveGaussianNoise(scale=0.1 * 255, per_channel=True),
iaa.Dropout(p=(0, 0.2)),
# iaa.WithChannels(0, iaa.Affine(rotate=(0, 0))),
iaa.ChannelShuffle(0.35),
iaa.WithColorspace(to_colorspace="HSV",
from_colorspace="RGB",
per_channel=0.5),
]),
# Dropout some pixels
iaa.OneOf([
iaa.Dropout((0.01, 0.1), per_channel=0.5),
iaa.CoarseDropout((0.01, 0.1),
size_percent=(0.01, 0.5),
per_channel=0.5),
iaa.SaltAndPepper((0.01, 0.3))
]),
# Play with the colors of the image
iaa.OneOf([
iaa.Invert(0.01, per_channel=0.5),
iaa.AddToHueAndSaturation((-1, 1)),
iaa.MultiplyHueAndSaturation((-1, 1))
]),
# Change brightness and contrast
iaa.OneOf([
iaa.Add((-10, 10), per_channel=0.5),
iaa.Multiply((0.5, 1.5), per_channel=0.5),
iaa.GammaContrast(gamma=(0.5, 1.75), per_channel=0.5),
iaa.SigmoidContrast(cutoff=(0, 1), per_channel=0.5),
iaa.LogContrast(gain=(0.5, 1), per_channel=0.5),
iaa.LinearContrast(alpha=(0.25, 1.75), per_channel=0.5),
iaa.HistogramEqualization()
]),
sometimes(iaa.ElasticTransformation(alpha=(0.5, 3.5),
sigma=0.25)),
# move pixels locally around (with random strengths)
sometimes(iaa.PiecewiseAffine(scale=(0.01, 0.05))),
sometimes = lambda aug: iaa.Sometimes(0.5, aug)
AUGMENTATIONS = iaa.Sequential([
iaa.Fliplr(0.5),
iaa.Flipud(0.5),
sometimes(iaa.Affine(
scale=(0.8, 1.2),
rotate=(90),
mode=ia.ALL)),
sometimes(iaa.ElasticTransformation(alpha=(0.8, 1.2),\
sigma=(9.0, 11.0))),
sometimes(iaa.AdditiveGaussianNoise(scale=(0, 0.1))),
sometimes(iaa.GaussianBlur((0, 0.1))),
sometimes(iaa.MultiplyBrightness((0.65, 1.35))),
sometimes(iaa.LinearContrast((0.5, 1.5))),
sometimes(iaa.MultiplyHueAndSaturation((-1, 1)))
], random_order=True)
# instantiate datagen objects
train_datagen = ImageDataAugmentor(
# featurewise_center=True,
# featurewise_std_normalization=True,
augment=AUGMENTATIONS,
rescale=1. / 255,
preprocess_input=None)
val_datagen = ImageDataAugmentor(rescale=1. / 255)
# define the ImageNet mean subtraction (in RGB order)
mean = np.array([123.68, 116.779, 103.939], dtype="float32")
# set the mean subtraction value for each of the data augmentation objects
def __getitem__(self, index):
mosaic_pro = random.random()
if mosaic_pro > 0:
img_id = self.images[index]
img, labels = self.load_mosaic(index)
all_ann = []
for da_label in labels:
da_label = da_label.tolist()
for da_l in da_label:
all_ann.append(da_l)
num_objs = min(len(all_ann), self.max_objs)
else:
positive_aug = random.random()
if positive_aug > 2:
index1 = random.randint(0, self.num_samples - 1)
# chartlet_dir = "/home/raid5/daming/HandDataMix/TrainImg/AnnImgMix"
img_id = self.images[index]
img_id1 = self.images[index1]
file_name = self.coco.loadImgs(ids=[img_id])[0]['file_name']
file_name1 = self.coco.loadImgs(ids=[img_id1])[0]['file_name']
path_num = random.random()
img_path = os.path.join(self.img_dir, file_name)
img_path1 = os.path.join(self.img_dir, file_name1)
# if path_num > 0.5:
# img_path = os.path.join(chartlet_dir, file_name)
ann_ids = self.coco.getAnnIds(imgIds=[img_id])
ann_ids1 = self.coco.getAnnIds(imgIds=[img_id1])
anns = self.coco.loadAnns(ids=ann_ids)
anns1 = self.coco.loadAnns(ids=ann_ids1)
img = cv2.imread(img_path)
img1 = cv2.imread(img_path1)
hand_num = len(anns1)
if hand_num > 0:
for ann1 in anns1:
ran_id = random.randint(0, 26000)
hand_x = ann1['bbox'][0]
hand_y = ann1['bbox'][1]
hand_w = ann1['bbox'][2]
hand_h = ann1['bbox'][3]
temp = img1[hand_y:hand_y + hand_h,
hand_x:hand_x + hand_w]
temp_h, temp_w, c = temp.shape
src_h, src_w, src_c = img.shape
for n in range(100):
min_src = min(src_w, src_h)
max_temp = max(temp_h, temp_w)
if (max_temp > 0.5 * min_src):
break
if (src_w < temp_w or src_h < temp_h):
break
x_tmp = random.randint(0, src_w - temp_w)
y_tmp = random.randint(0, src_h - temp_h)
src_rect = [
x_tmp, y_tmp, x_tmp + temp_w, y_tmp + temp_h
]
iou_all = 0
for gt in anns:
gt = [
gt['bbox'][0], gt['bbox'][1],
gt['bbox'][0] + gt['bbox'][2],
gt['bbox'][1] + gt['bbox'][3]
]
iou = self.compute_iou(gt, src_rect)
iou_all = iou_all + iou
# print(iou_all)
if iou_all == 0:
img[y_tmp:y_tmp + temp_h,
x_tmp:x_tmp + temp_w] = temp
a = {
'bbox': [x_tmp, y_tmp, temp_w, temp_h],
'category_id': 1
}
anns.append(a)
break
num_objs = min(len(anns), self.max_objs)
else:
img_id = self.images[index]
file_name = self.coco.loadImgs(ids=[img_id])[0]['file_name']
# daming_dir = "/home/raid5/daming/HandDataMix/TrainImg/AnnImgMix"
img_path = os.path.join(self.img_dir, file_name)
# img_path1 = os.path.join(daming_dir, file_name)
ann_ids = self.coco.getAnnIds(imgIds=[img_id])
anns = self.coco.loadAnns(ids=ann_ids)
num_objs = min(len(anns), self.max_objs)
img = cv2.imread(img_path)
# daming_num = random.random()
# if daming_num > 0.5:
# img = cv2.imread(img_path)
# else:
# img = cv2.imread(img_path1)
gray_pro = random.random()
if gray_pro > 2:
img = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
img = cv2.cvtColor(img, cv2.COLOR_GRAY2BGR)
height, width = img.shape[0], img.shape[1]
c = np.array([img.shape[1] / 2., img.shape[0] / 2.], dtype=np.float32)
if self.opt.keep_res:
input_h = (height | self.opt.pad) + 1
input_w = (width | self.opt.pad) + 1
s = np.array([input_w, input_h], dtype=np.float32)
else:
s = max(img.shape[0], img.shape[1]) * 1.0
input_h, input_w = self.opt.input_h, self.opt.input_w
flipped = False
if self.split == 'train':
if not self.opt.not_rand_crop:
s = s * np.random.choice(np.arange(0.6, 1.4, 0.1))
# s = s * np.random.choice(np.arange(0.3, 1.2, 0.1))
w_border = self._get_border(128, img.shape[1])
h_border = self._get_border(128, img.shape[0])
c[0] = np.random.randint(low=w_border,
high=img.shape[1] - w_border)
c[1] = np.random.randint(low=h_border,
high=img.shape[0] - h_border)
else:
sf = self.opt.scale
cf = self.opt.shift
c[0] += s * np.clip(np.random.randn() * cf, -2 * cf, 2 * cf)
c[1] += s * np.clip(np.random.randn() * cf, -2 * cf, 2 * cf)
s = s * np.clip(np.random.randn() * sf + 1, 1 - sf, 1 + sf)
if np.random.random() < self.opt.flip:
flipped = True
img = img[:, ::-1, :]
c[0] = width - c[0] - 1
trans_input = get_affine_transform(c, s, 0, [input_w, input_h])
inp = cv2.warpAffine(img,
trans_input, (input_w, input_h),
flags=cv2.INTER_LINEAR)
iaa_pro = random.random()
if iaa_pro > 2:
aug_seq = iaa.Sequential(
[iaa.MultiplyHueAndSaturation((0.5, 1.5), per_channel=True)])
# aug_seq = iaa.Sequential([
# iaa.Sometimes(
# 0.5,
# iaa.GaussianBlur(sigma=(0, 0.5))
# ),
# iaa.LinearContrast((0.75, 1.5)),
# iaa.AdditiveGaussianNoise(loc=0, scale=(0.0, 0.05*255), per_channel=0.5),
# iaa.Multiply((0.8, 1.2), per_channel=0.2),
# ], random_order=True)
inp, _ = aug_seq(image=inp, bounding_boxes=None)
inp = (inp.astype(np.float32) / 255.)
if self.split == 'train' and not self.opt.no_color_aug:
color_aug(self._data_rng, inp, self._eig_val, self._eig_vec)
inp = (inp - self.mean) / self.std
inp = inp.transpose(2, 0, 1)
output_h = input_h // self.opt.down_ratio
output_w = input_w // self.opt.down_ratio
num_classes = self.num_classes
trans_output = get_affine_transform(c, s, 0, [output_w, output_h])
hm = np.zeros((num_classes, output_h, output_w), dtype=np.float32)
wh = np.zeros((self.max_objs, 2), dtype=np.float32)
dense_wh = np.zeros((2, output_h, output_w), dtype=np.float32)
# ind is the center index, reg is the offset of center point in extracted feature maps
reg = np.zeros((self.max_objs, 2), dtype=np.float32)
ind = np.zeros((self.max_objs), dtype=np.int64)
reg_mask = np.zeros((self.max_objs), dtype=np.uint8)
cat_spec_wh = np.zeros((self.max_objs, num_classes * 2),
dtype=np.float32)
cat_spec_mask = np.zeros((self.max_objs, num_classes * 2),
dtype=np.uint8)
draw_gaussian = draw_msra_gaussian if self.opt.mse_loss else \
draw_umich_gaussian
gt_det = []
for k in range(num_objs):
if mosaic_pro > 0:
ann = all_ann[k]
bbox = np.array([
float(ann[0]),
float(ann[1]),
float(ann[2]),
float(ann[3])
],
dtype=np.float32)
else:
ann = anns[k]
bbox = self._coco_box_to_bbox(ann['bbox'])
cls_id = 0
if flipped:
bbox[[0, 2]] = width - bbox[[2, 0]] - 1
bbox[:2] = affine_transform(bbox[:2], trans_output)
bbox[2:] = affine_transform(bbox[2:], trans_output)
bbox[[0, 2]] = np.clip(bbox[[0, 2]], 0, output_w - 1)
bbox[[1, 3]] = np.clip(bbox[[1, 3]], 0, output_h - 1)
h, w = bbox[3] - bbox[1], bbox[2] - bbox[0]
if h > 0 and w > 0:
# print("- h : ", h," - w : ", w)
radius = gaussian_radius((math.ceil(h), math.ceil(w)))
radius = max(0, int(radius))
radius = self.opt.hm_gauss if self.opt.mse_loss else radius
ct = np.array([(bbox[0] + bbox[2]) / 2,
(bbox[1] + bbox[3]) / 2],
dtype=np.float32)
ct_int = ct.astype(np.int32)
draw_gaussian(hm[cls_id], ct_int, radius)
wh[k] = 1. * w, 1. * h
ind[k] = ct_int[1] * output_w + ct_int[0]
reg[k] = ct - ct_int
reg_mask[k] = 1
cat_spec_wh[k, cls_id * 2:cls_id * 2 + 2] = wh[k]
cat_spec_mask[k, cls_id * 2:cls_id * 2 + 2] = 1
if self.opt.dense_wh:
draw_dense_reg(dense_wh, hm.max(axis=0), ct_int, wh[k],
radius)
gt_det.append([
ct[0] - w / 2, ct[1] - h / 2, ct[0] + w / 2, ct[1] + h / 2,
1, cls_id
])
ret = {
'input': inp,
'hm': hm,
'reg_mask': reg_mask,
'ind': ind,
'wh': wh
}
if self.opt.dense_wh:
hm_a = hm.max(axis=0, keepdims=True)
dense_wh_mask = np.concatenate([hm_a, hm_a], axis=0)
ret.update({'dense_wh': dense_wh, 'dense_wh_mask': dense_wh_mask})
del ret['wh']
elif self.opt.cat_spec_wh:
ret.update({
'cat_spec_wh': cat_spec_wh,
'cat_spec_mask': cat_spec_mask
})
del ret['wh']
if self.opt.reg_offset:
ret.update({'reg': reg})
if self.opt.debug > 0 or not self.split == 'train':
gt_det = np.array(gt_det, dtype=np.float32) if len(gt_det) > 0 else \
np.zeros((1, 6), dtype=np.float32)
meta = {'c': c, 's': s, 'gt_det': gt_det, 'img_id': img_id}
ret['meta'] = meta
return ret
import imgaug.augmenters as iaa
import random
import numpy as np
import cv2
from PIL import Image
aug_transform = iaa.SomeOf((0, None), [
iaa.OneOf([
iaa.MultiplyAndAddToBrightness(mul=(0.3, 1.6), add=(-50, 50)),
iaa.MultiplyHueAndSaturation((0.5, 1.5), per_channel=True),
iaa.ChannelShuffle(0.5),
iaa.RemoveSaturation(),
iaa.Grayscale(alpha=(0.0, 1.0)),
iaa.ChangeColorTemperature((1100, 35000)),
]),
iaa.OneOf([
iaa.MedianBlur(k=(3, 7)),
iaa.BilateralBlur(
d=(3, 10), sigma_color=(10, 250), sigma_space=(10, 250)),
iaa.MotionBlur(k=(3, 9), angle=[-45, 45]),
iaa.MeanShiftBlur(spatial_radius=(5.0, 10.0),
color_radius=(5.0, 10.0)),
iaa.AllChannelsCLAHE(clip_limit=(1, 10)),
iaa.AllChannelsHistogramEqualization(),
iaa.GammaContrast((0.5, 1.5), per_channel=True),
iaa.GammaContrast((0.5, 1.5)),
iaa.SigmoidContrast(gain=(3, 10), cutoff=(0.4, 0.6), per_channel=True),
iaa.SigmoidContrast(gain=(3, 10), cutoff=(0.4, 0.6)),
iaa.HistogramEqualization(),
iaa.Sharpen(alpha=0.5)
color_augmentation = args.color
geometric_augmentation = args.geometric
all_aumenters = args.all
augmenters = [
iaa.Dropout(p=(0, 0.1)),
iaa.CoarseDropout((0.01, 0.05), size_percent=0.1),
iaa.Multiply((0.5, 1.3), per_channel=(0.2)),
iaa.GaussianBlur(sigma=(0, 5)),
iaa.AdditiveGaussianNoise(scale=((0, 0.2*255))),
iaa.ContrastNormalization((0.5, 1.5)),
iaa.Grayscale(alpha=((0.1, 1))),
iaa.ElasticTransformation(alpha=(0, 5.0), sigma=0.25),
iaa.PerspectiveTransform(scale=(0.15)),
iaa.MultiplyHueAndSaturation((0.7)),
iaa.Affine(scale=((0.6, 1.2))),
iaa.Affine(translate_percent=(-0.3, 0.3)),
iaa.Affine(shear=(-25, 25)),
iaa.Affine(translate_percent={"x": (-0.3, 0.3), "y": (-0.2, 0.2)}),
iaa.Fliplr(1),
iaa.Affine(scale={"x": (0.6, 1.4), "y": (0.6, 1.4)})
]
if all_aumenters > 0:
some_of_all = iaa.SomeOf(all_aumenters, augmenters)
# pixel manipulation
color_augmenters = [
iaa.Dropout(p=(0, 0.1)),
def create_augmenters(height, width, height_augmentable, width_augmentable,
only_augmenters):
def lambda_func_images(images, random_state, parents, hooks):
return images
def lambda_func_heatmaps(heatmaps, random_state, parents, hooks):
return heatmaps
def lambda_func_keypoints(keypoints, random_state, parents, hooks):
return keypoints
def assertlambda_func_images(images, random_state, parents, hooks):
return True
def assertlambda_func_heatmaps(heatmaps, random_state, parents, hooks):
return True
def assertlambda_func_keypoints(keypoints, random_state, parents, hooks):
return True
augmenters_meta = [
iaa.Sequential([iaa.Noop(), iaa.Noop()],
random_order=False,
name="Sequential_2xNoop"),
iaa.Sequential([iaa.Noop(), iaa.Noop()],
random_order=True,
name="Sequential_2xNoop_random_order"),
iaa.SomeOf((1, 3),
[iaa.Noop(), iaa.Noop(), iaa.Noop()],
random_order=False,
name="SomeOf_3xNoop"),
iaa.SomeOf((1, 3),
[iaa.Noop(), iaa.Noop(), iaa.Noop()],
random_order=True,
name="SomeOf_3xNoop_random_order"),
iaa.OneOf([iaa.Noop(), iaa.Noop(), iaa.Noop()], name="OneOf_3xNoop"),
iaa.Sometimes(0.5, iaa.Noop(), name="Sometimes_Noop"),
iaa.WithChannels([1, 2], iaa.Noop(), name="WithChannels_1_and_2_Noop"),
iaa.Noop(name="Noop"),
iaa.Lambda(func_images=lambda_func_images,
func_heatmaps=lambda_func_heatmaps,
func_keypoints=lambda_func_keypoints,
name="Lambda"),
iaa.AssertLambda(func_images=assertlambda_func_images,
func_heatmaps=assertlambda_func_heatmaps,
func_keypoints=assertlambda_func_keypoints,
name="AssertLambda"),
iaa.AssertShape((None, height_augmentable, width_augmentable, None),
name="AssertShape"),
iaa.ChannelShuffle(0.5, name="ChannelShuffle")
]
augmenters_arithmetic = [
iaa.Add((-10, 10), name="Add"),
iaa.AddElementwise((-10, 10), name="AddElementwise"),
#iaa.AddElementwise((-500, 500), name="AddElementwise"),
iaa.AdditiveGaussianNoise(scale=(5, 10), name="AdditiveGaussianNoise"),
iaa.AdditiveLaplaceNoise(scale=(5, 10), name="AdditiveLaplaceNoise"),
iaa.AdditivePoissonNoise(lam=(1, 5), name="AdditivePoissonNoise"),
iaa.Multiply((0.5, 1.5), name="Multiply"),
iaa.MultiplyElementwise((0.5, 1.5), name="MultiplyElementwise"),
iaa.Dropout((0.01, 0.05), name="Dropout"),
iaa.CoarseDropout((0.01, 0.05),
size_percent=(0.01, 0.1),
name="CoarseDropout"),
iaa.ReplaceElementwise((0.01, 0.05), (0, 255),
name="ReplaceElementwise"),
#iaa.ReplaceElementwise((0.95, 0.99), (0, 255), name="ReplaceElementwise"),
iaa.SaltAndPepper((0.01, 0.05), name="SaltAndPepper"),
iaa.ImpulseNoise((0.01, 0.05), name="ImpulseNoise"),
iaa.CoarseSaltAndPepper((0.01, 0.05),
size_percent=(0.01, 0.1),
name="CoarseSaltAndPepper"),
iaa.Salt((0.01, 0.05), name="Salt"),
iaa.CoarseSalt((0.01, 0.05),
size_percent=(0.01, 0.1),
name="CoarseSalt"),
iaa.Pepper((0.01, 0.05), name="Pepper"),
iaa.CoarsePepper((0.01, 0.05),
size_percent=(0.01, 0.1),
name="CoarsePepper"),
iaa.Invert(0.1, name="Invert"),
# ContrastNormalization
iaa.JpegCompression((50, 99), name="JpegCompression")
]
augmenters_blend = [
iaa.Alpha((0.01, 0.99), iaa.Noop(), name="Alpha"),
iaa.AlphaElementwise((0.01, 0.99), iaa.Noop(),
name="AlphaElementwise"),
iaa.SimplexNoiseAlpha(iaa.Noop(), name="SimplexNoiseAlpha"),
iaa.FrequencyNoiseAlpha((-2.0, 2.0),
iaa.Noop(),
name="FrequencyNoiseAlpha")
]
augmenters_blur = [
iaa.GaussianBlur(sigma=(1.0, 5.0), name="GaussianBlur"),
iaa.AverageBlur(k=(3, 11), name="AverageBlur"),
iaa.MedianBlur(k=(3, 11), name="MedianBlur"),
iaa.BilateralBlur(d=(3, 11), name="BilateralBlur"),
iaa.MotionBlur(k=(3, 11), name="MotionBlur")
]
augmenters_color = [
# InColorspace (deprecated)
iaa.WithColorspace(to_colorspace="HSV",
children=iaa.Noop(),
name="WithColorspace"),
iaa.WithHueAndSaturation(children=iaa.Noop(),
name="WithHueAndSaturation"),
iaa.MultiplyHueAndSaturation((0.8, 1.2),
name="MultiplyHueAndSaturation"),
iaa.MultiplyHue((-1.0, 1.0), name="MultiplyHue"),
iaa.MultiplySaturation((0.8, 1.2), name="MultiplySaturation"),
iaa.AddToHueAndSaturation((-10, 10), name="AddToHueAndSaturation"),
iaa.AddToHue((-10, 10), name="AddToHue"),
iaa.AddToSaturation((-10, 10), name="AddToSaturation"),
iaa.ChangeColorspace(to_colorspace="HSV", name="ChangeColorspace"),
iaa.Grayscale((0.01, 0.99), name="Grayscale"),
iaa.KMeansColorQuantization((2, 16), name="KMeansColorQuantization"),
iaa.UniformColorQuantization((2, 16), name="UniformColorQuantization")
]
augmenters_contrast = [
iaa.GammaContrast(gamma=(0.5, 2.0), name="GammaContrast"),
iaa.SigmoidContrast(gain=(5, 20),
cutoff=(0.25, 0.75),
name="SigmoidContrast"),
iaa.LogContrast(gain=(0.7, 1.0), name="LogContrast"),
iaa.LinearContrast((0.5, 1.5), name="LinearContrast"),
iaa.AllChannelsCLAHE(clip_limit=(2, 10),
tile_grid_size_px=(3, 11),
name="AllChannelsCLAHE"),
iaa.CLAHE(clip_limit=(2, 10),
tile_grid_size_px=(3, 11),
to_colorspace="HSV",
name="CLAHE"),
iaa.AllChannelsHistogramEqualization(
name="AllChannelsHistogramEqualization"),
iaa.HistogramEqualization(to_colorspace="HSV",
name="HistogramEqualization"),
]
augmenters_convolutional = [
iaa.Convolve(np.float32([[0, 0, 0], [0, 1, 0], [0, 0, 0]]),
name="Convolve_3x3"),
iaa.Sharpen(alpha=(0.01, 0.99), lightness=(0.5, 2), name="Sharpen"),
iaa.Emboss(alpha=(0.01, 0.99), strength=(0, 2), name="Emboss"),
iaa.EdgeDetect(alpha=(0.01, 0.99), name="EdgeDetect"),
iaa.DirectedEdgeDetect(alpha=(0.01, 0.99), name="DirectedEdgeDetect")
]
augmenters_edges = [iaa.Canny(alpha=(0.01, 0.99), name="Canny")]
augmenters_flip = [
iaa.Fliplr(1.0, name="Fliplr"),
iaa.Flipud(1.0, name="Flipud")
]
augmenters_geometric = [
iaa.Affine(scale=(0.9, 1.1),
translate_percent={
"x": (-0.05, 0.05),
"y": (-0.05, 0.05)
},
rotate=(-10, 10),
shear=(-10, 10),
order=0,
mode="constant",
cval=(0, 255),
name="Affine_order_0_constant"),
iaa.Affine(scale=(0.9, 1.1),
translate_percent={
"x": (-0.05, 0.05),
"y": (-0.05, 0.05)
},
rotate=(-10, 10),
shear=(-10, 10),
order=1,
mode="constant",
cval=(0, 255),
name="Affine_order_1_constant"),
iaa.Affine(scale=(0.9, 1.1),
translate_percent={
"x": (-0.05, 0.05),
"y": (-0.05, 0.05)
},
rotate=(-10, 10),
shear=(-10, 10),
order=3,
mode="constant",
cval=(0, 255),
name="Affine_order_3_constant"),
iaa.Affine(scale=(0.9, 1.1),
translate_percent={
"x": (-0.05, 0.05),
"y": (-0.05, 0.05)
},
rotate=(-10, 10),
shear=(-10, 10),
order=1,
mode="edge",
cval=(0, 255),
name="Affine_order_1_edge"),
iaa.Affine(scale=(0.9, 1.1),
translate_percent={
"x": (-0.05, 0.05),
"y": (-0.05, 0.05)
},
rotate=(-10, 10),
shear=(-10, 10),
order=1,
mode="constant",
cval=(0, 255),
backend="skimage",
name="Affine_order_1_constant_skimage"),
# TODO AffineCv2
iaa.PiecewiseAffine(scale=(0.01, 0.05),
nb_rows=4,
nb_cols=4,
order=1,
mode="constant",
name="PiecewiseAffine_4x4_order_1_constant"),
iaa.PiecewiseAffine(scale=(0.01, 0.05),
nb_rows=4,
nb_cols=4,
order=0,
mode="constant",
name="PiecewiseAffine_4x4_order_0_constant"),
iaa.PiecewiseAffine(scale=(0.01, 0.05),
nb_rows=4,
nb_cols=4,
order=1,
mode="edge",
name="PiecewiseAffine_4x4_order_1_edge"),
iaa.PiecewiseAffine(scale=(0.01, 0.05),
nb_rows=8,
nb_cols=8,
order=1,
mode="constant",
name="PiecewiseAffine_8x8_order_1_constant"),
iaa.PerspectiveTransform(scale=(0.01, 0.05),
keep_size=False,
name="PerspectiveTransform"),
iaa.PerspectiveTransform(scale=(0.01, 0.05),
keep_size=True,
name="PerspectiveTransform_keep_size"),
iaa.ElasticTransformation(
alpha=(1, 10),
sigma=(0.5, 1.5),
order=0,
mode="constant",
cval=0,
name="ElasticTransformation_order_0_constant"),
iaa.ElasticTransformation(
alpha=(1, 10),
sigma=(0.5, 1.5),
order=1,
mode="constant",
cval=0,
name="ElasticTransformation_order_1_constant"),
iaa.ElasticTransformation(
alpha=(1, 10),
sigma=(0.5, 1.5),
order=1,
mode="nearest",
cval=0,
name="ElasticTransformation_order_1_nearest"),
iaa.ElasticTransformation(
alpha=(1, 10),
sigma=(0.5, 1.5),
order=1,
mode="reflect",
cval=0,
name="ElasticTransformation_order_1_reflect"),
iaa.Rot90((1, 3), keep_size=False, name="Rot90"),
iaa.Rot90((1, 3), keep_size=True, name="Rot90_keep_size")
]
augmenters_pooling = [
iaa.AveragePooling(kernel_size=(1, 16),
keep_size=False,
name="AveragePooling"),
iaa.AveragePooling(kernel_size=(1, 16),
keep_size=True,
name="AveragePooling_keep_size"),
iaa.MaxPooling(kernel_size=(1, 16), keep_size=False,
name="MaxPooling"),
iaa.MaxPooling(kernel_size=(1, 16),
keep_size=True,
name="MaxPooling_keep_size"),
iaa.MinPooling(kernel_size=(1, 16), keep_size=False,
name="MinPooling"),
iaa.MinPooling(kernel_size=(1, 16),
keep_size=True,
name="MinPooling_keep_size"),
iaa.MedianPooling(kernel_size=(1, 16),
keep_size=False,
name="MedianPooling"),
iaa.MedianPooling(kernel_size=(1, 16),
keep_size=True,
name="MedianPooling_keep_size")
]
augmenters_segmentation = [
iaa.Superpixels(p_replace=(0.05, 1.0),
n_segments=(10, 100),
max_size=64,
interpolation="cubic",
name="Superpixels_max_size_64_cubic"),
iaa.Superpixels(p_replace=(0.05, 1.0),
n_segments=(10, 100),
max_size=64,
interpolation="linear",
name="Superpixels_max_size_64_linear"),
iaa.Superpixels(p_replace=(0.05, 1.0),
n_segments=(10, 100),
max_size=128,
interpolation="linear",
name="Superpixels_max_size_128_linear"),
iaa.Superpixels(p_replace=(0.05, 1.0),
n_segments=(10, 100),
max_size=224,
interpolation="linear",
name="Superpixels_max_size_224_linear"),
iaa.UniformVoronoi(n_points=(250, 1000), name="UniformVoronoi"),
iaa.RegularGridVoronoi(n_rows=(16, 31),
n_cols=(16, 31),
name="RegularGridVoronoi"),
iaa.RelativeRegularGridVoronoi(n_rows_frac=(0.07, 0.14),
n_cols_frac=(0.07, 0.14),
name="RelativeRegularGridVoronoi"),
]
augmenters_size = [
iaa.Resize((0.8, 1.2), interpolation="nearest", name="Resize_nearest"),
iaa.Resize((0.8, 1.2), interpolation="linear", name="Resize_linear"),
iaa.Resize((0.8, 1.2), interpolation="cubic", name="Resize_cubic"),
iaa.CropAndPad(percent=(-0.2, 0.2),
pad_mode="constant",
pad_cval=(0, 255),
keep_size=False,
name="CropAndPad"),
iaa.CropAndPad(percent=(-0.2, 0.2),
pad_mode="edge",
pad_cval=(0, 255),
keep_size=False,
name="CropAndPad_edge"),
iaa.CropAndPad(percent=(-0.2, 0.2),
pad_mode="constant",
pad_cval=(0, 255),
name="CropAndPad_keep_size"),
iaa.Pad(percent=(0.05, 0.2),
pad_mode="constant",
pad_cval=(0, 255),
keep_size=False,
name="Pad"),
iaa.Pad(percent=(0.05, 0.2),
pad_mode="edge",
pad_cval=(0, 255),
keep_size=False,
name="Pad_edge"),
iaa.Pad(percent=(0.05, 0.2),
pad_mode="constant",
pad_cval=(0, 255),
name="Pad_keep_size"),
iaa.Crop(percent=(0.05, 0.2), keep_size=False, name="Crop"),
iaa.Crop(percent=(0.05, 0.2), name="Crop_keep_size"),
iaa.PadToFixedSize(width=width + 10,
height=height + 10,
pad_mode="constant",
pad_cval=(0, 255),
name="PadToFixedSize"),
iaa.CropToFixedSize(width=width - 10,
height=height - 10,
name="CropToFixedSize"),
iaa.KeepSizeByResize(iaa.CropToFixedSize(height=height - 10,
width=width - 10),
interpolation="nearest",
name="KeepSizeByResize_CropToFixedSize_nearest"),
iaa.KeepSizeByResize(iaa.CropToFixedSize(height=height - 10,
width=width - 10),
interpolation="linear",
name="KeepSizeByResize_CropToFixedSize_linear"),
iaa.KeepSizeByResize(iaa.CropToFixedSize(height=height - 10,
width=width - 10),
interpolation="cubic",
name="KeepSizeByResize_CropToFixedSize_cubic"),
]
augmenters_weather = [
iaa.FastSnowyLandscape(lightness_threshold=(100, 255),
lightness_multiplier=(1.0, 4.0),
name="FastSnowyLandscape"),
iaa.Clouds(name="Clouds"),
iaa.Fog(name="Fog"),
iaa.CloudLayer(intensity_mean=(196, 255),
intensity_freq_exponent=(-2.5, -2.0),
intensity_coarse_scale=10,
alpha_min=0,
alpha_multiplier=(0.25, 0.75),
alpha_size_px_max=(2, 8),
alpha_freq_exponent=(-2.5, -2.0),
sparsity=(0.8, 1.0),
density_multiplier=(0.5, 1.0),
name="CloudLayer"),
iaa.Snowflakes(name="Snowflakes"),
iaa.SnowflakesLayer(density=(0.005, 0.075),
density_uniformity=(0.3, 0.9),
flake_size=(0.2, 0.7),
flake_size_uniformity=(0.4, 0.8),
angle=(-30, 30),
speed=(0.007, 0.03),
blur_sigma_fraction=(0.0001, 0.001),
name="SnowflakesLayer")
]
augmenters = (augmenters_meta + augmenters_arithmetic + augmenters_blend +
augmenters_blur + augmenters_color + augmenters_contrast +
augmenters_convolutional + augmenters_edges +
augmenters_flip + augmenters_geometric + augmenters_pooling +
augmenters_segmentation + augmenters_size +
augmenters_weather)
if only_augmenters is not None:
augmenters_reduced = []
for augmenter in augmenters:
if any([
re.search(pattern, augmenter.name)
for pattern in only_augmenters
]):
augmenters_reduced.append(augmenter)
augmenters = augmenters_reduced
return augmenters
def simple_imgaug_example():
image_dir_path = dataset_home_dir_path + '/phenotyping/cvppp2017_lsc_lcc_challenge/package/CVPPP2017_LSC_training/training/A1'
label_dir_path = dataset_home_dir_path + '/phenotyping/cvppp2017_lsc_lcc_challenge/package/CVPPP2017_LSC_training/training/A1'
images, labels = prepare_dataset(image_dir_path, label_dir_path)
image_width, image_height = 200, 200
# FIXME [decide] >> Before or after random transformation?
# Preprocessing (normalization, standardization, etc).
images_pp = images.astype(np.float)
#images_pp /= 255.0
images_pp = standardize_samplewise(images_pp)
#images_pp = standardize_featurewise(images_pp)
if True:
augmenter = iaa.SomeOf(
(1, 2),
[
iaa.OneOf([
iaa.Affine(
scale={
'x': (0.8, 1.2),
'y': (0.8, 1.2)
}, # Scale images to 80-120% of their size, individually per axis.
translate_percent={
'x': (-0.1, 0.1),
'y': (-0.1, 0.1)
}, # Translate by -10 to +10 percent (per axis).
rotate=(-10, 10), # Rotate by -10 to +10 degrees.
shear=(-5, 5), # Shear by -5 to +5 degrees.
#order=[0, 1], # Use nearest neighbour or bilinear interpolation (fast).
order=
0, # Use nearest neighbour or bilinear interpolation (fast).
#cval=(0, 255), # If mode is constant, use a cval between 0 and 255.
#mode=ia.ALL # Use any of scikit-image's warping modes (see 2nd image from the top for examples).
#mode='edge' # Use any of scikit-image's warping modes (see 2nd image from the top for examples).
),
#iaa.PiecewiseAffine(scale=(0.01, 0.05)), # Move parts of the image around. Slow.
iaa.PerspectiveTransform(scale=(0.01, 0.1)),
iaa.ElasticTransformation(
alpha=(20.0, 50.0), sigma=(6.5, 8.5)
), # Move pixels locally around (with random strengths).
]),
iaa.OneOf([
iaa.GaussianBlur(sigma=(
0,
3.0)), # Blur images with a sigma between 0 and 3.0.
iaa.AverageBlur(
k=(2, 7)
), # Blur image using local means with kernel sizes between 2 and 7.
iaa.MedianBlur(
k=(3, 11)
), # Blur image using local medians with kernel sizes between 2 and 7.
iaa.MotionBlur(k=(5, 11),
angle=(0, 360),
direction=(-1.0, 1.0),
order=1),
]),
iaa.OneOf([
iaa.AdditiveGaussianNoise(
loc=0, scale=(0.1 * 255, 0.5 * 255),
per_channel=False), # Add Gaussian noise to images.
iaa.AdditiveLaplaceNoise(loc=0,
scale=(0.1 * 255, 0.4 * 255),
per_channel=False),
iaa.AdditivePoissonNoise(lam=(32, 96), per_channel=False),
iaa.CoarseSaltAndPepper(p=(0.1, 0.3),
size_percent=(0.2, 0.9),
per_channel=False),
iaa.CoarseSalt(p=(0.1, 0.3),
size_percent=(0.2, 0.9),
per_channel=False),
iaa.CoarsePepper(p=(0.1, 0.3),
size_percent=(0.2, 0.9),
per_channel=False),
iaa.CoarseDropout(p=(0.1, 0.3),
size_percent=(0.05, 0.3),
per_channel=False),
]),
iaa.OneOf([
iaa.MultiplyHueAndSaturation(mul=(-10, 10),
per_channel=False),
iaa.AddToHueAndSaturation(value=(-255, 255),
per_channel=False),
iaa.LinearContrast(
alpha=(0.5, 1.5),
per_channel=False), # Improve or worsen the contrast.
iaa.Invert(p=1,
per_channel=False), # Invert color channels.
iaa.Sharpen(alpha=(0, 1.0),
lightness=(0.75, 1.5)), # Sharpen images.
iaa.Emboss(alpha=(0, 1.0),
strength=(0, 2.0)), # Emboss images.
]),
],
random_order=True)
elif False:
augmenter = iaa.Sequential(
[
# Apply the following augmenters to most images.
iaa.Fliplr(0.5), # Horizontally flip 50% of all images.
iaa.Flipud(0.2), # Vertically flip 20% of all images.
# Crop images by -5% to 10% of their height/width.
iaa.Sometimes(
0.5,
iaa.CropAndPad(percent=(-0.05, 0.1),
pad_mode=ia.ALL,
pad_cval=(0, 255))),
iaa.Sometimes(
0.5,
iaa.Affine(
scale={
'x': (0.8, 1.2),
'y': (0.8, 1.2)
}, # Scale images to 80-120% of their size, individually per axis.
translate_percent={
'x': (-0.2, 0.2),
'y': (-0.2, 0.2)
}, # Translate by -20 to +20 percent (per axis).
rotate=(-45, 45), # Rotate by -45 to +45 degrees.
shear=(-16, 16), # Shear by -16 to +16 degrees.
order=[
0,
1
], # Use nearest neighbour or bilinear interpolation (fast).
cval=(
0, 255
), # If mode is constant, use a cval between 0 and 255.
mode=ia.
ALL # Use any of scikit-image's warping modes (see 2nd image from the top for examples).
)),
# Execute 0 to 5 of the following (less important) augmenters per image.
# Don't execute all of them, as that would often be way too strong.
iaa.SomeOf(
(0, 5),
[
iaa.Sometimes(
0.5,
iaa.Superpixels(p_replace=(0, 1.0),
n_segments=(20, 200))
), # Convert images into their superpixel representation.
iaa.OneOf([
iaa.GaussianBlur(
(0, 3.0)
), # Blur images with a sigma between 0 and 3.0.
iaa.AverageBlur(
k=(2, 7)
), # Blur image using local means with kernel sizes between 2 and 7.
iaa.MedianBlur(
k=(3, 11)
), # Blur image using local medians with kernel sizes between 2 and 7.
]),
iaa.Sharpen(alpha=(0, 1.0),
lightness=(0.75, 1.5)), # Sharpen images.
iaa.Emboss(alpha=(0, 1.0),
strength=(0, 2.0)), # Emboss images.
# Search either for all edges or for directed edges, blend the result with the original image using a blobby mask.
iaa.SimplexNoiseAlpha(
iaa.OneOf([
iaa.EdgeDetect(alpha=(0.5, 1.0)),
iaa.DirectedEdgeDetect(alpha=(0.5, 1.0),
direction=(0.0, 1.0)),
])),
iaa.AdditiveGaussianNoise(
loc=0, scale=(0.0, 0.05 * 255),
per_channel=0.5), # Add gaussian noise to images.
iaa.OneOf([
iaa.Dropout(
(0.01, 0.1), per_channel=0.5
), # Randomly remove up to 10% of the pixels.
iaa.CoarseDropout((0.03, 0.15),
size_percent=(0.02, 0.05),
per_channel=0.2),
]),
iaa.Invert(0.05,
per_channel=True), # Invert color channels.
iaa.Add(
(-10, 10), per_channel=0.5
), # Change brightness of images (by -10 to 10 of original value).
iaa.AddToHueAndSaturation(
(-20, 20)), # Change hue and saturation.
# Either change the brightness of the whole image (sometimes per channel) or change the brightness of subareas.
iaa.OneOf([
iaa.Multiply((0.5, 1.5), per_channel=0.5),
iaa.FrequencyNoiseAlpha(
exponent=(-4, 0),
first=iaa.Multiply(
(0.5, 1.5), per_channel=True),
second=iaa.ContrastNormalization((0.5, 2.0)))
]),
iaa.ContrastNormalization(
(0.5, 2.0), per_channel=0.5
), # Improve or worsen the contrast.
iaa.Grayscale(alpha=(0.0, 1.0)),
iaa.Sometimes(
0.5,
iaa.ElasticTransformation(alpha=(0.5, 3.5),
sigma=0.25)
), # Move pixels locally around (with random strengths).
iaa.Sometimes(
0.5, iaa.PiecewiseAffine(scale=(0.01, 0.05))
), # Sometimes move parts of the image around.
iaa.Sometimes(
0.5, iaa.PerspectiveTransform(scale=(0.01, 0.1)))
],
random_order=True)
],
random_order=True)
else:
augmenter = iaa.Sequential([
iaa.SomeOf(
1,
[
#iaa.Sometimes(0.5, iaa.Crop(px=(0, 100))), # Crop images from each side by 0 to 16px (randomly chosen).
iaa.Sometimes(0.5, iaa.Crop(percent=(
0,
0.1))), # Crop images by 0-10% of their height/width.
iaa.Fliplr(0.5), # Horizontally flip 50% of the images.
iaa.Flipud(0.5), # Vertically flip 50% of the images.
iaa.Sometimes(
0.5,
iaa.Affine(
scale={
'x': (0.8, 1.2),
'y': (0.8, 1.2)
}, # Scale images to 80-120% of their size, individually per axis.
translate_percent={
'x': (-0.2, 0.2),
'y': (-0.2, 0.2)
}, # Translate by -20 to +20 percent (per axis).
rotate=(-45, 45), # Rotate by -45 to +45 degrees.
shear=(-16, 16), # Shear by -16 to +16 degrees.
#order=[0, 1], # Use nearest neighbour or bilinear interpolation (fast).
order=
0, # Use nearest neighbour or bilinear interpolation (fast).
#cval=(0, 255), # If mode is constant, use a cval between 0 and 255.
#mode=ia.ALL # Use any of scikit-image's warping modes (see 2nd image from the top for examples).
#mode='edge' # Use any of scikit-image's warping modes (see 2nd image from the top for examples).
)),
iaa.Sometimes(0.5, iaa.GaussianBlur(
sigma=(0,
3.0))) # Blur images with a sigma of 0 to 3.0.
]),
iaa.Scale(size={
'height': image_height,
'width': image_width
}) # Resize.
])
for idx in range(images.shape[0]):
images_pp[idx] = (images_pp[idx] - np.min(images_pp[idx])) / (
np.max(images_pp[idx]) - np.min(images_pp[idx])) * 255
images_pp = images_pp.astype(np.uint8)
# Test 1 (good).
augmenter_det = augmenter.to_deterministic(
) # Call this for each batch again, NOT only once at the start.
#images_aug1 = augmenter_det.augment_images(images)
images_aug1 = augmenter_det.augment_images(images_pp)
labels_aug1 = augmenter_det.augment_images(labels)
augmenter_det = augmenter.to_deterministic(
) # Call this for each batch again, NOT only once at the start.
#images_aug2 = augmenter_det.augment_images(images)
images_aug2 = augmenter_det.augment_images(images_pp)
labels_aug2 = augmenter_det.augment_images(labels)
#export_images(images, labels, './augmented1/img', '')
export_images(images_pp, labels, './augmented1/img', '')
export_images(images_aug1, labels_aug1, './augmented1/img', '_aug1')
export_images(images_aug2, labels_aug2, './augmented1/img', '_aug2')
# Test 2 (bad).
augmenter_det = augmenter.to_deterministic(
) # Call this for each batch again, NOT only once at the start.
#images_aug1 = augmenter_det.augment_images(images)
images_aug1 = augmenter_det.augment_images(images_pp)
labels_aug1 = augmenter_det.augment_images(labels)
#images_aug2 = augmenter_det.augment_images(images)
images_aug2 = augmenter_det.augment_images(images_pp)
labels_aug2 = augmenter_det.augment_images(labels)
#export_images(images, labels, './augmented2/img', '')
export_images(images_pp, labels, './augmented2/img', '')
export_images(images_aug1, labels_aug1, './augmented2/img', '_aug1')
export_images(images_aug2, labels_aug2, './augmented2/img', '_aug2')
print('*********************************', images_pp.dtype)
rotate=(-20, 20), # rotate by -45 to +45 degrees
shear=(-16, 16), # shear by -16 to +16 degrees
order=[
0, 1
], # use nearest neighbour or bilinear interpolation (fast)
mode=
"edge" # use any of scikit-image"s warping modes (see 2nd image from the top for examples)
)),
sometimes(iaa.PiecewiseAffine(
scale=(0.01, 0.05))), # sometimes move parts of the image around
iaa.PerspectiveTransform(scale=(0.01, 0.1)),
sometimes(
iaa.OneOf([
iaa.SigmoidContrast(gain=5),
iaa.Multiply(mul=1.0),
iaa.MultiplyHueAndSaturation(mul=1.25)
])),
],
random_order=True)
def apply_batch(images, file_counter_start, aug_amount):
i = file_counter_start
batches = []
for _ in range(aug_amount):
batches.append(iaa.UnnormalizedBatch(images=images))
for batch in seq.augment_batches(batches=batches, background=True):
for images in batch.images_aug:
cv2.imwrite(f"augmented/{category}/{category}_{i}.png", images)
i += 1