Working of the image smoothing with median method in OpenCV

Recipe Objective: How does the image smoothing work with the median method in OpenCV?

In this recipe, let us understand what image smoothing is and how it works with the Median Smoothing method.

Step 1: Import the libraries and read the image.

Let us first import the necessary libraries and read the image. The image that we are using here is the one shown below.

 

import numpy as np
import cv2
from matplotlib import pyplot as plt
image = cv2.imread('projectpro_noise_50.jpg')


Step 2: Image smoothing / Image blurring using Median Smoothing

Image smoothing is a noise elimination technique used to remove unwanted dots and disturbances in the image. As we can see, the image that we have is boisterous. This noise can be eliminated using the Image smoothing technique. There are various ways to perform Image Smoothing, and Median Smoothing is one of them.

In median soothing, we calculate the median of all the pixels under the kernel area and replace the central element with the median. A kernel is nothing but a small matrix used for sharpening, blurring, embossing, edge detection, and much more. It is also sometimes called a convolution matrix, a mask, or a filter. Median smoothing is highly effective in eliminating salt-and-pepper noise ( Salt-and-pepper noise, sometimes called impulse noise, is the discrepancies caused in the image due to sudden or sharp disturbances. The best example for such a noisy image is the input image that we have ) in any image. Another reason median smoothing works effectively is because it does not replace the central element with any new value. Instead it replaces the central element with one of the values already present inside the kernel area and hence it reduces noise effectively.

Median smoothing can be implemented in OpenCV using cv2.medianBlur() function which takes the following parameters

• src: The image which is to be smoothened
• ksize: The kernel dimension

The kernel dimension is a positive odd integer value.

median = cv2.medianBlur(image,5)

Step 3: Displaying the output

It's time to see and understand how the noise has been eliminated from our image. Let us use matplotlib subplots to display the input and the output image and analyze them.

titles = ['Original Image',"Median Smoothing"]
images = [image,median]
plt.figure(figsize=(13,5))
for i in range(2):
    plt.subplot(1,2,i+1)
    plt.imshow(cv2.cvtColor(images[i], cv2.COLOR_BGR2RGB))
    plt.title(titles[i])
    plt.xticks([])
    plt.yticks([])
plt.tight_layout()
plt.show()

Output:

We can clearly see from the above output that the output has less noise and sharpness.