A library for Data Augmentation of images in computer vision.
Project description
🖼️💥 Augmently :
An Open Source library for Data Augmentation for image classification.
With Flips, Square crops and resiszing, and Salt and Pepper Noise.
📁Note On Folder Format:
Currently the library works only for image grouped each in their own class folder.
For example:
──My_animal_images_folder
├── Dogs
│ ├── dog_img_1.jpg
│ ├── dog_img_2.jpg
│ ├── ...
│ └── dog_img_n.jpeg
├── Cats
├── ...
└── Elephants
⭐Start Using it:
Download the Augmently folder and place it in your project folder. COPY PASTE the code below:
from Augmently.Augmently import create_resized_cropped_square_class_images, create_salt_and_pepper_class_images, create_salt_and_pepper_class_images
🤖 The Functions:
🔲 Crop to Square Size - create_resized_cropped_square_class_images()
What it does:
Creates a new folder with your images both resized and cropped to the square image length of your choice
Arguments:
image_data_path (String) , output_path (string), square_length (Number)
Example Usage:
create_resized_cropped_square_class_images_in_new_folder("Desktop/image_folder", "Desktop/image_folder_square_resized_224pixels", 224 )
🧂Salt and Pepper Noise - create_salt_and_pepper_class_images()
What it does:
Creates a new folder with your images with your desired amount of salt and pepper noise pixels added to your images
Arguments:
image_data_path (String) , output_path (String) , noise_amount (Number)
Example Usage:
create_salt_and_pepper_class_images_in_new_folder("Desktop/image_folder", "Desktop/image_folder_salt_pepper_0.05", 0.05 )
↔️ Flip images - create_flipped_class_images()
What it does:
Creates a new folder with your images flipped
Arguments:
image_data_path (String) , output_path (string)
Example Usage:
create_flipped_class_images_in_new_folder("Desktop/image_folder", "Desktop/image_folder_flipped")
🔜 In Progress:
🔄 Square Cropped Rotations - create_rotated_class_images()
What it does:
Creates a new folder with your images with your desired amount of maximum rotation to your images
Arguments:
image_data_path (String) , output_path (String) , max_rotation (Number in degrees)
Example Usage:
create_rotated_images_in_new_folder("Desktop/image_folder", "Desktop/image_folder_rotated_360", 360 )
🌈 Gaussian Noise - create_gaussian_noise_class_images()
What it does:
Creates a new folder with your images with your desired amount of gaussian noise to your images
Arguments:
image_data_path (String) , output_path (String) , amount_noise (Number)
Example Usage:
create_gaussian_noise_images_in_new_folder("Desktop/image_folder", "Desktop/image_gaussian_0.05", 0.05 )
⬛ Add Occluding Black Squares - create_occluding_squares_class_images()
What it does:
Creates a new folder with your images with your desired max size of occluding black squares to your images
Arguments:
image_data_path (String) , output_path (String) , max_square_len (Number in px)
Example Usage:
create_occluding_square_images_in_new_folder("Desktop/image_folder", "Desktop/image_black_square_20", 20 )
🔍 Random Resized Zooms - create_zoom_class_images()
What it does:
Creates a new folder with your images with your desired max zoom (resized to original len) to your images
Arguments:
image_data_path (String) , output_path (String) , max_zoom_square_len (Number in px)
Example Usage:
create_zoomed_images_in_new_folder("Desktop/image_folder", "Desktop/image_zoom_120", 120 )
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
File details
Details for the file Augmently-1.0.9.tar.gz
.
File metadata
- Download URL: Augmently-1.0.9.tar.gz
- Upload date:
- Size: 4.3 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/2.0.0 pkginfo/1.5.0.1 requests/2.21.0 setuptools/41.6.0 requests-toolbelt/0.9.1 tqdm/4.31.1 CPython/3.7.3
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | 43372c2c220506ebb78179aa3cf19a73a3ec7f5a0ab7eaeba66b77188cc37df4 |
|
MD5 | 48c53fad034012b9715460f31339802f |
|
BLAKE2b-256 | d476728301bc3e6c3b3fdf82a075ac28a37db404c4720b5d39e1cfb2ced9affe |