Image Processing Library

This is a Python library for performing common image processing operations such as brightness adjustment, contrast adjustment, blurring, and image combination.

Features

• Brightness Adjustment: Adjust the brightness of an image by scaling the pixel intensities.
• Contrast Adjustment: Adjust the contrast of an image relative to a specified midpoint.
• Blurring: Apply a box filter blur to an image with proper edge handling.
• Kernel Application: Apply a custom convolution kernel to an image with improved edge handling.
• Image Combination: Combine two images using various methods (root sum of squares, addition, multiplication, maximum).
• Utility Functions: Create Gaussian and edge detection kernels.

Installation

To use this library, you'll need to have the following dependencies installed:

• NumPy

You can install the dependencies using pip:

pip install numpy

Usage

Here's an example of how to use the library:

from image import Image
from image_processing import brighten, adjust_contrast, blur, apply_kernel, combine_images

# Load an image
image = Image(filename='example.jpg')

# Brighten the image
brightened_image = brighten(image, factor=1.5)

# Adjust the contrast
contrast_adjusted_image = adjust_contrast(image, factor=2.0, mid=0.5)

# Apply a Gaussian blur
blurred_image = blur(image, kernel_size=5)

# Apply a custom kernel
edge_kernel = create_edge_kernel('sobel_x')
edged_image = apply_kernel(image, edge_kernel)

# Combine two images
image1 = Image(filename='image1.jpg')
image2 = Image(filename='image2.jpg')
combined_image = combine_images(image1, image2, method='root_sum_square')

API Reference

Functions

brighten(image: Image, factor: float) -> Image

Brighten or darken the image by scaling each pixel's intensity.

adjust_contrast(image: Image, factor: float, mid: float = 0.5) -> Image

Adjust contrast relative to a midpoint using vectorized operations.

blur(image: Image, kernel_size: int) -> Image

Blur image using an optimized box filter approach with proper edge handling.

apply_kernel(image: Image, kernel: np.ndarray, padding_mode: str = 'edge') -> Image

Apply a convolution kernel to the image with improved edge handling.

combine_images(image1: Image, image2: Image, method: str = 'root_sum_square') -> Image

Combine two images using various methods.

create_gaussian_kernel(size: int, sigma: float) -> np.ndarray

Create a Gaussian kernel for blurring.

create_edge_kernel(direction: str) -> np.ndarray

Create kernels for edge detection.

Contributing

If you find any issues or have suggestions for improvements, please feel free to open an issue or submit a pull request.

License

This project is licensed under the MIT License.