Python Pillow - Removing Noise

Removing noise, also referred to as denoising, involves the process of reducing unwanted artifacts in an image. Noise in an image typically appears as random variations in brightness or color that are not part of the original scene or subject being photographed. The goal of image denoising is to enhance the quality of an image by eliminating or reducing these unwanted and distracting artifacts, making the image cleaner, and more visually appealing.

The Python pillow library offers a range of denoising filters, allowing users to remove noise from noisy images and recover the original image. In this tutorial, we will explore GaussianBlur and Median filters as effective methods for noise removal.

Removing the noise using the GaussianBlur filter

Removing the noise from an image using the gaussian blur filter is a widely used technique. This technique is works by applying a convolution filter to the image to smooth out the pixel values.

Example - Using Gaussian Blur Filter

Here is an example that uses the ImageFilter.GaussianBlur() filter to remove the noise from an RGB image.

main.py

from PIL import Image, ImageFilter

# Open a Gaussian Noised Image  
input_image = Image.open("Images/GaussianNoisedImage.jpg").convert('RGB')

# Apply Gaussian blur with some radius
blurred_image = input_image.filter(ImageFilter.GaussianBlur(radius=2))

# Display the input and the blurred image
input_image.show()
blurred_image.show()

Input Noised Image

Output Noise Removed Image

Example - Removing the noise using the Median Filter

The median filter is an alternative approach for noise reduction, particularly useful for the images where the noise points are like small and scattered. This function operates by replacing each pixel value with the median value within its local neighborhood. the Pillow library provides the ImageFilter.MedianFilter() filter for this purpose.

Following example removes the noice from an image using the ImageFilter.MedianFilter().

main.py

from PIL import Image, ImageFilter

# Open an image
input_image = Image.open("Images/balloons_noisy.jpg")

# Apply median filter with a kernel size 
filtered_image = input_image.filter(ImageFilter.MedianFilter(size=3))

# Display the input and the filtered image
input_image.show()
filtered_image.show()

Input Noisy Image

Output Median Filtered Image

Exampl - Reducing Noise in Grayscale Image

This example demonstrates the reduction of salt-and-pepper noise in a grayscale image. This can be done by using a median filter, enhancing the contrast to improve visibility, and subsequently converting the image to binary mode for display.

main.py

from PIL import Image, ImageEnhance, ImageFilter, ImageOps

# Open the image and convert it to grayscale
input_image = ImageOps.grayscale(Image.open('Images/salt-and-pepper_noise.jpg'))

# Apply a median filter with a kernel size of 5 to reduce noise
filtered_image = input_image.filter(ImageFilter.MedianFilter(5))

# Enhance the contrast of the filtered image
contrast_enhancer = ImageEnhance.Contrast(filtered_image)
high_contrast_image = contrast_enhancer.enhance(3)

# Convert the image to binary
binary_image = high_contrast_image.convert('1')

# Display the input and processed images
input_image.show()
binary_image.show()

Input Noisy Image

Output Median Filtered Image