How to do it...

Perform the following steps:

1. Import the packages:

import cv2
import numpy as np
import matplotlib.pyplot as plt

2. Load an image, convert it to floating-point, and scale it down to the [0, 1] range:

image = cv2.imread('../data/Lena.png').astype(np.float32) / 255

3. Create noise in the image by adding random values to each pixel, and display it: 

noised = (image + 0.2 * np.random.rand(*image.shape).astype(np.float32))
noised = noised.clip(0, 1)
plt.imshow(noised[:,:,[2,1,0]])
plt.show()

4. Apply GaussianBlur to the noisy image and show the result:

gauss_blur = cv2.GaussianBlur(noised, (7, 7), 0)
plt.imshow(gauss_blur[:, :, [2, 1, 0]])
plt.show()

5. Apply median filtering:

median_blur = cv2.medianBlur((noised * 255).astype(np.uint8), 7)
plt.imshow(median_blur[:, :, [2, 1, 0]])
plt.show()

6. Perform median filtration to our image with noise:

bilat = cv2.bilateralFilter(noised, -1, 0.3, 10)
plt.imshow(bilat[:, :, [2, 1, 0]])
plt.show()