Scikit Image - Masking an Image

Quiz

Masking an image is a fundamental technique in image processing and it's used for a wide range of tasks such as region of interest selection, image segmentation, and selective filtering. Applying a mask to an image, effectively "cover-ups" or "reveal" specified parts of the image based on the mask's regions. This allows you to perform various operations on the selected regions while leaving the rest of the image unaffected.

The scikit-image (skimage) library provides functions like flood() and flood_fill() that enable users to perform flood fill operations on images.

Using the skimage.morphology.flood() function

The morphology.flood() function is used for flood filling regions in an n-dimensional array. It starts from a specified seed point and fills connected points that are equal to or within a certain tolerance of the seed value found.

Syntax

Following is the syntax of this function −

skimage.morphology.flood(image, seed_point, *, footprint=None, connectivity=None, tolerance=None)

Parameters

image (ndarray): An n-dimensional array on which flood fill will be performed.
seed_point (tuple or int): The starting point for the flood fill. If the image is 1D, you can specify this point as an integer.
footprint (ndarray, optional): This is the footprint (structuring element) used to determine the neighborhood of each evaluated pixel. It should contain only 1's and 0's and must have the same number of dimensions as the image. If not provided, all adjacent pixels are considered part of the neighborhood (fully connected).
connectivity (int, optional): A number used to determine the neighborhood of each evaluated pixel. Pixels whose squared distance from the center is less than or equal to connectivity are considered neighbors. This parameter is ignored if footprint is not None.
tolerance (float or int, optional): If set to None (default), adjacent values must be strictly equal to the initial value of the image at the seed_point. If a value is provided, a comparison will be done at every point, and if the value is within tolerance of the initial value, it will also be filled (inclusive).

Return Value

It returns a Boolean array with the same shape as the input image. It has True values for areas connected to and equal (or within tolerance of) the seed point and False for all other values.

Example

The following example demonstrates how to use the morphology.flood() function to perform a flood fill operation.

import numpy as np
from skimage.morphology import flood
import matplotlib.pyplot as plt

# Create an example image
image = np.zeros((4, 7), dtype=int)
image[1:3, 1:3] = 1
image[3, 0] = 1
image[1:3, 4:6] = 2
image[3, 6] = 3

# Perform flood fill starting from (1, 1)
mask = flood(image, (1, 1), connectivity=1)
image_flooded = image.copy()
image_flooded[mask] = 5

# Plot the original and thinned images
fig, axes = plt.subplots(1, 2, figsize=(10, 5))
ax = axes.ravel()

# Display the original binary image
ax[0].imshow(image, cmap=plt.cm.gray)
ax[0].axis('off')
ax[0].set_title('Original Image')

# Display the Thinned image
ax[1].imshow(image_flooded, cmap=plt.cm.gray)
ax[1].axis('off')
ax[1].set_title('Flooded Image')

plt.tight_layout()
plt.show()

Output

On executing the above program, you will get the following output −

Using the skimage.morphology.flood_fill() function

The morphology.flood_fill() function is used to perform flood filling on an image. It starts at a specified seed point and fills connected points that are equal to or within a certain tolerance of the seed value with a new specified value.