DirectoryNoiseAdder Module

The DirectoryNoiseAdder class is an extension of the abstract base class NoiseAdder. It provides functionality for applying noise to multiple images located in a directory and saving the resulting noisy images to a specified output path. This class is ideal for batch image processing tasks where noise needs to be added to a collection of images.

Attributes

• dir_path : str

  The path to the directory that contains the images to be processed.

• output_path : str

  The path where the processed images will be saved.

  Note

  If left empty, the images will be saved in the same directory as the source images.

• noises : list[Noise]

  A list of Noise objects that represent different noise types to be applied to the images. Each Noise object should adhere to the structure and behavior defined by the Noise class from the iftg.noises module.

• identifier : str

  A unique identifier appended to the filenames of processed images. This is useful for distinguishing between original and processed images (e.g., an image named example.tif could become example_noisy.tif).

• img_formats : list[str]

  A list of image formats in which the processed images will be saved. These formats are typically strings like 'TIF', 'JPEG', etc.

Methods

_apply_noises()

_apply_noises(self, image: Image) -> tuple[Image.Image, str, str]

This method applies the specified noise transformations to a given image.

• Parameters:

  • image : Image

    The image object to which noises will be applied.

• Returns:

  • A tuple containing:
    • The transformed Image object (after applying noise),
    • The base name of the image file (without its extension),
    • The image file extension (with a dot, e.g., .tif).

add_noises()

add_noises(self) -> list[tuple[Image.Image, str, str]]

This method processes all the images in the directory by applying the specified noises.

• Returns:

  • A list[tuple], where each tuple contains:
    • A noisy Image.Image,
    • The image's base name str,
    • The image's format or extension str.

add_noises()

add_noises(self) -> list[tuple[Image.Image, str, str]]

This method processes all the images in the directory by applying the specified noises.

• Returns:

  • A list[tuple], where each tuple contains:
    • A noisy Image.Image,
    • The image's base name str,
    • The image's format or extension str.

save_image()

save_image(self, img_info: tuple[Image.Image, str, str]) -> None

This method saves a noisy image to the output path.

• Parameters:

  • img_info : tuple[Image.Image, str, str]

    A tuple containing the noisy image, the base name of the image, and the image format.

transform_images()

transform_images(self) -> None

This method triggers the entire process of transforming and saving images by applying noise. It is abstract and should be implemented by subclasses to manage the full transformation workflow.

• Returns:
  • None

Usage Example

from iftg.noises import GaussianNoise
from iftg.adders import DirectoryNoiseAdder

# Define the noises to be applied
noises = [GaussianNoise(mean=0, sigma=25)]

# Create an instance of DirectoryNoiseAdder
dir_noise_adder = DirectoryNoiseAdder(
    dir_path='path/to/input_images',
    output_path='path/to/output_images',
    noises=noises,
    identifier='gaussian_noisy',
    img_formats=['jpg', 'png']
)

# Transform and save noisy images
dir_noise_adder.transform_images()

In this example, a list of Noise objects is created (in this case, a GaussianNoise object), and the DirectoryNoiseAdder class is instantiated with the paths to input and output directories. It will apply the noise to all images in the directory and save the resulting noisy images.

Notes

1. Image Formats:

   The class supports reading and saving images in any format listed in the img_formats attribute, making it flexible for various image file types.

2. Error Handling:

   If the dir_path does not exist, the class raises a FileNotFoundError. Additionally, if no output_path is provided, the processed images will be saved in the original dir_path.

3. Efficiency:

   The use of Python's reduce function allows for efficient chaining of noise transformations on each image. The class processes images in the order they are found, and can handle large directories by iterating over the images one by one.