Morphological operations (erosion, dilation) in SciPy - Mini Project: Build & Apply

Morphological Operations with SciPy

📖 Scenario: Imagine you have a simple black and white image represented as a grid of 0s and 1s. You want to clean up the image by removing small noise or filling small holes. Morphological operations like erosion and dilation help with this.

🎯 Goal: You will create a small binary image using a NumPy array, set a structuring element, apply erosion and dilation using SciPy, and then display the results.

📋 What You'll Learn

Create a binary image as a 2D NumPy array with exact values

Create a structuring element using SciPy's morphology module

Apply erosion and dilation to the image using the structuring element

Print the original image and the results of erosion and dilation

💡 Why This Matters

🌍 Real World

Morphological operations are used in image processing to clean up images, remove noise, and prepare images for further analysis like object detection.

💼 Career

Understanding and applying morphological operations is important for roles in computer vision, medical imaging, and any job involving image data preprocessing.

1

Create the binary image

Create a 2D NumPy array called image with these exact values: [[0, 0, 1, 1, 0], [0, 1, 1, 1, 0], [1, 1, 1, 1, 1], [0, 1, 1, 1, 0], [0, 0, 1, 1, 0]]

SciPy

# Create the binary image array called image
# Your code here

Need a hint?

Use np.array to create a 2D array with the exact rows and columns shown.

2

Create the structuring element

Import scipy.ndimage as ndi and create a 3x3 structuring element called structure using ndi.generate_binary_structure(2, 1)

SciPy

import numpy as np

image = np.array([[0, 0, 1, 1, 0],
                  [0, 1, 1, 1, 0],
                  [1, 1, 1, 1, 1],
                  [0, 1, 1, 1, 0],
                  [0, 0, 1, 1, 0]])

# Import scipy.ndimage as ndi and create a structuring element called structure
# Your code here

Need a hint?

Use ndi.generate_binary_structure with 2 dimensions and connectivity 1 to create a 3x3 structuring element.

3

Apply erosion and dilation

Use ndi.binary_erosion with image and structure to create eroded. Use ndi.binary_dilation with image and structure to create dilated.

SciPy

import numpy as np
import scipy.ndimage as ndi

image = np.array([[0, 0, 1, 1, 0],
                  [0, 1, 1, 1, 0],
                  [1, 1, 1, 1, 1],
                  [0, 1, 1, 1, 0],
                  [0, 0, 1, 1, 0]])

structure = ndi.generate_binary_structure(2, 1)

# Apply binary erosion and dilation to image using structure
# Your code here

Need a hint?

Use ndi.binary_erosion and ndi.binary_dilation functions with the structure parameter.

4

Print the original and processed images

Print the image, eroded, and dilated arrays each on separate lines using print().

SciPy

import numpy as np
import scipy.ndimage as ndi

image = np.array([[0, 0, 1, 1, 0],
                  [0, 1, 1, 1, 0],
                  [1, 1, 1, 1, 1],
                  [0, 1, 1, 1, 0],
                  [0, 0, 1, 1, 0]])

structure = ndi.generate_binary_structure(2, 1)

eroded = ndi.binary_erosion(image, structure=structure)
dilated = ndi.binary_dilation(image, structure=structure)

# Print image, eroded, and dilated arrays
# Your code here

Need a hint?

Use print() to show each array. The eroded and dilated arrays will show True and False values.