aug¶

Classes

Functions

Description

Augmentations for images and annotations

class RotationModes[source]¶

Bases: object

Enum-like class for rotation behavior: KEEP (pad to fit) or CROP.

crop(img, ann, top_pad=0, left_pad=0, bottom_pad=0, right_pad=0)[source]¶

Crops an Image and Annotation from all sides with a given values.

Parameters:

img : np.ndarray¶

RGB image in numpy format.

ann¶

Input annotation.

top_pad : int, optional¶

Top padding in pixels.

left_pad : int, optional¶

Left padding in pixels.

bottom_pad : int, optional¶

Bottom padding in pixels.

right_pad : int, optional¶

Right padding in pixels.

Raises:

RuntimeError – if image shape does not match img_size in annotation.

Returns:

Tuple containing cropped image and annotation.

Return type:

Tuple[np.ndarray, Annotation]

Usage Example:

import os
from dotenv import load_dotenv

import supervisely as sly
from supervisely.aug.aug import crop

# Load secrets and create API object from .env file (recommended)
# Learn more here: https://developer.supervisely.com/getting-started/basics-of-authentication
if sly.is_development():
    load_dotenv(os.path.expanduser("~/supervisely.env"))

api = sly.Api.from_env()

# Download image and annotation from API
project_id = 116501
image_id = 193940171

meta_json = api.project.get_meta(project_id)
meta = sly.ProjectMeta.from_json(meta_json)

image_np = api.image.download_np(image_id) # <class 'numpy.ndarray'>
print(image_np.shape)
# Output: (800, 1067, 3)

ann_info = api.annotation.download(image_id)
ann = sly.Annotation.from_json(ann_info.annotation, meta)

crop_image_np, crop_ann = crop(image_np, ann, top_pad=50, left_pad=100, bottom_pad=50, right_pad=100)
print(crop_image_np.shape)
# Output: (700, 867, 3)

crop_fraction(img, ann, top=0, left=0, bottom=0, right=0)[source]¶

Crops an Image and Annotation from all sides with the given fraction values.

Parameters:

img : np.ndarray¶

RGB image in numpy format.

ann¶

Input annotation.

top : int, optional¶

Top padding in pixels.

left : int, optional¶

Left padding in pixels.

bottom : int, optional¶

Bottom padding in pixels.

right : int, optional¶

Right padding in pixels.

Raises:

ValueError – if fraction values not between 0 and 1

Returns:

Tuple containing cropped image and annotation.

Return type:

Tuple[np.ndarray, Annotation]

Usage Example:

import os
from dotenv import load_dotenv

import supervisely as sly
from supervisely.aug.aug import crop_fraction

# Load secrets and create API object from .env file (recommended)
# Learn more here: https://developer.supervisely.com/getting-started/basics-of-authentication
if sly.is_development():
    load_dotenv(os.path.expanduser("~/supervisely.env"))

api = sly.Api.from_env()

# Download image and annotation from API
project_id = 116501
image_id = 193940171

meta_json = api.project.get_meta(project_id)
meta = sly.ProjectMeta.from_json(meta_json)

image_np = api.image.download_np(image_id) # <class 'numpy.ndarray'>
print(image_np.shape)
# Output: (800, 1067, 3)

ann_info = api.annotation.download(image_id)
ann = sly.Annotation.from_json(ann_info.annotation, meta)

crop_image_np, crop_ann = crop_fraction(image_np, ann, top=0.1, left=0.2, bottom=0.1, right=0.2)
print(crop_image_np.shape)
# Output: (640, 641, 3)

fliplr(img, ann)[source]¶

Flips an Image and Annotation around vertical axis.

Parameters:

img : np.ndarray¶

RGB image in numpy format.

ann¶

Input annotation.

Raises:

RuntimeError – if image shape does not match img_size in annotation.

Returns:

Tuple containing flipped image and annotation.

Return type:

Tuple[np.ndarray, Annotation]

Usage Example:

import os
from dotenv import load_dotenv

import supervisely as sly
from supervisely.aug.aug import fliplr

# Load secrets and create API object from .env file (recommended)
# Learn more here: https://developer.supervisely.com/getting-started/basics-of-authentication
if sly.is_development():
    load_dotenv(os.path.expanduser("~/supervisely.env"))

api = sly.Api.from_env()

# Download image and annotation from API
project_id = 116501
image_id = 193940171

meta_json = api.project.get_meta(project_id)
meta = sly.ProjectMeta.from_json(meta_json)

image_np = api.image.download_np(image_id) # <class 'numpy.ndarray'>
ann_info = api.annotation.download(image_id)
ann = sly.Annotation.from_json(ann_info.annotation, meta)

# Flip image and annotation
flip_image_np, flip_ann = fliplr(image_np, ann)

flipud(img, ann)[source]¶

Flips an Image and Annotation around horizontal axis.

Parameters:

img : np.ndarray¶

RGB image in numpy format.

ann¶

Input annotation.

Raises:

RuntimeError – if image shape does not match img_size in annotation.

Returns:

Tuple containing flipped image and annotation.

Return type:

Tuple[np.ndarray, Annotation]

Usage Example:

import os
from dotenv import load_dotenv

import supervisely as sly
from supervisely.aug.aug import flipud

# Load secrets and create API object from .env file (recommended)
# Learn more here: https://developer.supervisely.com/getting-started/basics-of-authentication
if sly.is_development():
    load_dotenv(os.path.expanduser("~/supervisely.env"))

api = sly.Api.from_env()

# Download image and annotation from API
project_id = 116501
image_id = 193940171

meta_json = api.project.get_meta(project_id)
meta = sly.ProjectMeta.from_json(meta_json)

image_np = api.image.download_np(image_id) # <class 'numpy.ndarray'>
ann_info = api.annotation.download(image_id)
ann = sly.Annotation.from_json(ann_info.annotation, meta)

# Flip image and annotation
flip_image_np, flip_ann = flipud(image_np, ann)

instance_crop(img, ann, class_title, save_other_classes_in_crop=True, padding_config=None)[source]¶

Crops objects of specified classes from Image and Annotation with configurable padding.

Parameters:

img : np.ndarray¶

RGB image in numpy format.

ann¶

Input annotation.

class_title : str¶

Name of class to crop.

save_other_classes_in_crop : bool, optional¶

If True saves non-target classes in each cropped Annotation, otherwise don’t.

padding_config : dict, optional¶

Dict with padding.

Raises:

ValueError – if padding size format is incorrect

Returns:

List of cropped (image numpy array, annotation) pairs

Return type:

List[Tuple[np.ndarray, Annotation]]

Usage Example:

import os
from dotenv import load_dotenv

import supervisely as sly
from supervisely.aug.aug import instance_crop

# Load secrets and create API object from .env file (recommended)
# Learn more here: https://developer.supervisely.com/getting-started/basics-of-authentication
if sly.is_development():
    load_dotenv(os.path.expanduser("~/supervisely.env"))

api = sly.Api.from_env()

# Download image and annotation from API
project_id = 116501
image_id = 193940171

meta_json = api.project.get_meta(project_id)
meta = sly.ProjectMeta.from_json(meta_json)

image_np = api.image.download_np(image_id) # <class 'numpy.ndarray'>
print(image_np.shape)
# Output: (800, 1067, 3)

ann_info = api.annotation.download(image_id)
ann = sly.Annotation.from_json(ann_info.annotation, meta)

result = instance_crop(image_np, ann, 'kiwi', True, {'top': '20px', 'left': '50px', 'bottom': '700px', 'right': '1000px'})
for crop_image_np, crop_ann in result:
    print(crop_image_np.shape)
# Output: (270, 635, 3)
#         (426, 345, 3)

random_crop(img, ann, height, width)[source]¶

Crops an Image and Annotation at a random location.

Parameters:

img : np.ndarray¶

RGB image in numpy format.

ann¶

Input annotation.

height : int, optional¶

Desired height of output crop.

width : int, optional¶

Desired width of output crop.

Raises:

RuntimeError – if image shape does not match img_size in annotation.

Returns:

Tuple containing cropped image and annotation.

Return type:

Tuple[np.ndarray, Annotation]

Usage Example:

import os
from dotenv import load_dotenv

import supervisely as sly
from supervisely.aug.aug import random_crop

# Load secrets and create API object from .env file (recommended)
# Learn more here: https://developer.supervisely.com/getting-started/basics-of-authentication
if sly.is_development():
    load_dotenv(os.path.expanduser("~/supervisely.env"))

api = sly.Api.from_env()

# Download image and annotation from API
project_id = 116501
image_id = 193940171

meta_json = api.project.get_meta(project_id)
meta = sly.ProjectMeta.from_json(meta_json)

image_np = api.image.download_np(image_id) # <class 'numpy.ndarray'>
print(image_np.shape)
# Output: (800, 1067, 3)

ann_info = api.annotation.download(image_id)
ann = sly.Annotation.from_json(ann_info.annotation, meta)

crop_image_np, crop_ann = random_crop(image_np, ann, height=500, width=700)
print(crop_image_np.shape)
# Output: (500, 700, 3)

random_crop_fraction(img, ann, height_fraction_range, width_fraction_range)[source]¶

Crops an Image and Annotation at a random location with random size in a given interval.

Parameters:

img : np.ndarray¶

RGB image in numpy format.

ann¶

Input annotation.

height_fraction_range : Tuple[float, float]¶

Range of relative values [0, 1] to select output height from.

width_fraction_range : Tuple[float, float]¶

Range of relative values [0, 1] to select output width from.

Raises:

RuntimeError – if image shape does not match img_size in annotation.

Returns:

Tuple containing cropped image and annotation.

Return type:

Tuple[np.ndarray, Annotation]

Usage Example:

import os
from dotenv import load_dotenv

import supervisely as sly
from supervisely.aug.aug import random_crop_fraction

# Load secrets and create API object from .env file (recommended)
# Learn more here: https://developer.supervisely.com/getting-started/basics-of-authentication
if sly.is_development():
    load_dotenv(os.path.expanduser("~/supervisely.env"))

api = sly.Api.from_env()

# Download image and annotation from API
project_id = 116501
image_id = 193940171

meta_json = api.project.get_meta(project_id)
meta = sly.ProjectMeta.from_json(meta_json)

image_np = api.image.download_np(image_id) # <class 'numpy.ndarray'>
print(image_np.shape)
# Output: (800, 1067, 3)

ann_info = api.annotation.download(image_id)
ann = sly.Annotation.from_json(ann_info.annotation, meta)

crop_image_np, crop_ann = random_crop_fraction(image_np, ann, height_fraction_range=(0.1, 0.8), width_fraction_range=(0.1, 0.8))
print(crop_image_np.shape)
# Output: (486, 585, 3)

resize(img, ann, size, skip_empty_masks=False)[source]¶

Resizes an input Image and Annotation to a given size.

Parameters:

img : np.ndarray¶

RGB image in numpy format.

ann¶

Input annotation.

size : Tuple[int, int]¶

Desired size (height, width) in pixels or -1.

skip_empty_masks : bool, optional¶

If True, skips resizing of empty masks in Annotation.

Raises:

RuntimeError – if image shape does not match img_size in annotation.

Returns:

Tuple containing resized image and annotation.

Return type:

Tuple[np.ndarray, Annotation]

Usage Example:

import os
from dotenv import load_dotenv

import supervisely as sly
from supervisely.aug.aug import resize

# Load secrets and create API object from .env file (recommended)
# Learn more here: https://developer.supervisely.com/getting-started/basics-of-authentication
if sly.is_development():
    load_dotenv(os.path.expanduser("~/supervisely.env"))

api = sly.Api.from_env()

# Download image and annotation from API
project_id = 116501
image_id = 193940171

meta_json = api.project.get_meta(project_id)
meta = sly.ProjectMeta.from_json(meta_json)

image_np = api.image.download_np(image_id) # <class 'numpy.ndarray'>
print(image_np.shape)
# Output: (800, 1067, 3)

ann_info = api.annotation.download(image_id)
ann = sly.Annotation.from_json(ann_info.annotation, meta)

resize_image_np, resize_ann = resize(image_np, ann, (600, -1))
print(resize_image_np.shape)
# Output: (600, 800, 3)

rotate(img, ann, degrees, mode='keep')[source]¶

Rotates an Image and Annotation by random angle.

Parameters:

img : np.ndarray¶

RGB image in numpy format.

ann¶

Input annotation.

degrees : int¶

Rotation angle.

mode='keep'¶

One of RotateMode enum values.

Raises:

RuntimeError – if image shape does not match img_size in annotation.

Returns:

Tuple containing rotated image and annotation.

Return type:

Tuple[np.ndarray, Annotation]

Usage Example:

import os
from dotenv import load_dotenv

import supervisely as sly
from supervisely.aug.aug import rotate

# Load secrets and create API object from .env file (recommended)
# Learn more here: https://developer.supervisely.com/getting-started/basics-of-authentication
if sly.is_development():
    load_dotenv(os.path.expanduser("~/supervisely.env"))

api = sly.Api.from_env()

# Download image and annotation from API
project_id = 116501
image_id = 193940171

meta_json = api.project.get_meta(project_id)
meta = sly.ProjectMeta.from_json(meta_json)

image_np = api.image.download_np(image_id) # <class 'numpy.ndarray'>
print(image_np.shape)
# Output: (800, 1067, 3)

ann_info = api.annotation.download(image_id)
ann = sly.Annotation.from_json(ann_info.annotation, meta)

rotate_image_np, rotate_ann = rotate(image_np, ann, 30)
print(rotate_image_np.shape)
# Output: (1231, 1326, 3)

scale(img, ann, frow=None, fcol=None, f=None)[source]¶

Scales an input Image and Annotation to a given size.

Parameters:

img : np.ndarray¶

RGB image in numpy format.

ann¶

Input annotation.

frow : float, optional¶

Desired height scale height value.

fcol : float, optional¶

Desired width scale height value.

f : float, optional¶

Desired height and width scale values in one(positive).

Raises:

RuntimeError – if image shape does not match img_size in annotation.

Returns:

Tuple containing scaled image and annotation.

Return type:

Tuple[np.ndarray, Annotation]

Usage Example:

import os
from dotenv import load_dotenv

import supervisely as sly
from supervisely.aug.aug import scale

# Load secrets and create API object from .env file (recommended)
# Learn more here: https://developer.supervisely.com/getting-started/basics-of-authentication
if sly.is_development():
    load_dotenv(os.path.expanduser("~/supervisely.env"))

api = sly.Api.from_env()

# Download image and annotation from API
project_id = 116501
image_id = 193940171

meta_json = api.project.get_meta(project_id)
meta = sly.ProjectMeta.from_json(meta_json)

image_np = api.image.download_np(image_id) # <class 'numpy.ndarray'>
print(image_np.shape)
# Output: (800, 1067, 3)

ann_info = api.annotation.download(193940171)
ann = sly.Annotation.from_json(ann_info.annotation, meta)

scale_image_np, scale_ann = scale(image_np, ann, frow=0.7, fcol=0.8)
print(scale_image_np.shape)
# Output: (560, 854, 3)