Notebooks

If you found errors in the notebooks, please create:

1. Pull request with the fix or
2. Submit an issue to the main repository of Albumentations

• example.ipynb. Defining a simple augmentation pipeline for image augmentation.
• example_bboxes.ipynb. Using Albumentations to augment bounding boxes for object detection tasks.
• example_bboxes2.ipynb. How to use Albumentations for detection tasks if you need to keep all bounding boxes.
• example_kaggle_salt.ipynb. Using Albumentations for a semantic segmentation task.
• example_keypoints.ipynb. Using Albumentations to augment keypoints.
• example_multi_target.ipynb. Applying the same augmentation with the same parameters to multiple images, masks, bounding boxes, or keypoints.
• example_weather_transforms.ipynb. Weather augmentations in Albumentations.
• migrating_from_torchvision_to_albumentations.ipynb. Migrating from torchvision to Albumentations.
• replay.ipynb. Debugging an augmentation pipeline with ReplayCompose.
• serialization.ipynb. How to save and load parameters of an augmentation pipeline.
• showcase.ipynb. Showcase. Cool augmentation examples on diverse set of images from various real-world tasks.
• pytorch_classification.ipynb. PyTorch and Albumentations for image classification.
• pytorch_semantic_segmentation.ipynb. PyTorch and Albumentations for semantic segmentation.
• xy_transform.ipynb. How to apply XYMasking.
• example_chromatic_aberration.ipynb. How to apply ChromaticAberration.
• example_d4.ipynb. How to apply D4 transform.
• example_gridshuffle.ipynb. How to apply RandomGridShuffle transform.
• example_documents.ipynb. How to apply Morphological, both erosion and dilation.
• example_hfhub.ipynb. How to load and save augmentation pipeline to HuggungFace Hub.
• example_OverlayElements.ipynb. How to apply OverlayElements to update text on the image
• example_textimage.ipynb. How to add text to the image

Usage examples

For detailed examples, see notebooks.

from albumentations import (
    Affine, CLAHE,Blur, OpticalDistortion, GridDistortion, HueSaturationValue,
    GaussNoise, MotionBlur, MedianBlur,
    RandomBrightnessContrast, D4, OneOf, Compose
)
import numpy as np

def strong_aug(p=0.5):
    return Compose([
        D4(),
        GaussNoise(),
        OneOf([
            MotionBlur(p=0.2),
            MedianBlur(blur_limit=3, p=0.1),
            Blur(blur_limit=3, p=0.1),
        ], p=0.2),
        Affine(translate_percent=0.0625, scale=(0.8, 1.2), rotate=(-45, 45), p=0.2),
        OneOf([
            OpticalDistortion(p=0.3),
            GridDistortion(p=0.1)
        ], p=0.2),
        OneOf([
            CLAHE(clip_limit=2),
            RandomBrightnessContrast(),
        ], p=0.3),
        HueSaturationValue(p=0.3),
    ], p=p, seed=137, strict=True)

image = np.ones((300, 300, 3), dtype=np.uint8)
mask = np.ones((300, 300), dtype=np.uint8)
whatever_data = "my name"
augmentation = strong_aug(p=0.9)
data = {"image": image, "mask": mask, "whatever_data": whatever_data, "additional": "hello"}
augmented = augmentation(**data)
image, mask, whatever_data, additional = augmented["image"], augmented["mask"], augmented["whatever_data"], augmented["additional"]

Augmentations examples

MultiplicativeNoise

1. Original image
2. MultiplicativeNoise(multiplier=0.5, p=1)
3. MultiplicativeNoise(multiplier=1.5, p=1)
4. MultiplicativeNoise(multiplier=[0.5, 1.5], per_channel=True, p=1)
5. MultiplicativeNoise(multiplier=[0.5, 1.5], elementwise=True, p=1)
6. MultiplicativeNoise(multiplier=[0.5, 1.5], elementwise=True, per_channel=True, p=1)

ToSepia

1. Original image
2. ToSepia(p=1)

ChannelDropout

1. Original image
2. ChannelDropout(channel_drop_range=(1, 1), fill_value=0, p=1)
3. ChannelDropout(channel_drop_range=(1, 1), fill_value=0, p=1)
4. ChannelDropout(channel_drop_range=(1, 1), fill_value=0, p=1)
5. ChannelDropout(channel_drop_range=(1, 1), fill_value=0, p=1)
6. ChannelDropout(channel_drop_range=(1, 1), fill_value=0, p=1)
7. ChannelDropout(channel_drop_range=(2, 2), fill_value=0, p=1)
8. ChannelDropout(channel_drop_range=(2, 2), fill_value=0, p=1)
9. ChannelDropout(channel_drop_range=(2, 2), fill_value=0, p=1)

ChannelShuffle

1. Original image
2. ChannelShuffle(p=1)
3. ChannelShuffle(p=1)
4. ChannelShuffle(p=1)

ChannelShuffle

1. Original image
2. ToGray(p=1)

InvertImg

1. Original image
2. InvertImg(p=1)

VerticalFlip

1. Original image
2. VerticalFlip(p=1)

HorizontalFlip

1. Original image
2. HorizontalFlip(p=1)

---

RandomGridShuffle

1. Original image
2. RandomGridShuffle(grid=(3, 3), p=1)
3. RandomGridShuffle(grid=(5, 5), p=1)
4. RandomGridShuffle(grid=(7, 7), p=1)

Blur

1. Original image
2. Blur(blur_limit=(7, 7), p=1)
3. Blur(blur_limit=(15, 15), p=1)
4. Blur(blur_limit=(50, 50), p=1)
5. Blur(blur_limit=(100, 100), p=1)
6. Blur(blur_limit=(300, 300), p=1)