Morphological operations (erosion, dilation) in SciPy - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to perform erosion on the binary image using scipy.

SciPy

from scipy import ndimage

binary_image = [[0, 1, 1, 0], [1, 1, 1, 0], [0, 1, 0, 0]]
eroded_image = ndimage.binary_[1](binary_image)
print(eroded_image)

Drag options to blanks, or click blank then click option'

Aerosion

Bdilation

Copening

Dclosing

Attempts:

3 left

💡 Hint

Common Mistakes

Using dilation instead of erosion

Misspelling the function name

Confusing opening and erosion

The binary_erosion function performs erosion on a binary image, shrinking the white regions.

2fill in blank

medium

Complete the code to perform dilation on the binary image using scipy.

SciPy

from scipy import ndimage

binary_image = [[0, 1, 0], [1, 0, 1], [0, 1, 0]]
dilated_image = ndimage.binary_[1](binary_image)
print(dilated_image)

Drag options to blanks, or click blank then click option'

Aerosion

Bdilation

Cgradient

Dhit_or_miss

Attempts:

3 left

💡 Hint

Common Mistakes

Using erosion instead of dilation

Using unrelated morphological functions

Forgetting to import ndimage

The binary_dilation function expands the white regions in a binary image.

3fill in blank

hard

Fix the error in the code to correctly perform erosion with a 3x3 structuring element.

SciPy

from scipy import ndimage
import numpy as np

binary_image = np.array([[1, 1, 0], [1, 0, 1], [0, 1, 1]])
structure = np.ones((3, 3))
eroded = ndimage.binary_[1](binary_image, structure=structure)
print(eroded)

Drag options to blanks, or click blank then click option'

Aopening

Bclosing

Cerosion

Ddilation

Attempts:

3 left

💡 Hint

Common Mistakes

Using dilation instead of erosion

Passing structuring element to wrong argument

Misspelling function name

Using binary_erosion with a 3x3 structuring element correctly erodes the image.

4fill in blank

hard

Fill both blanks to create a dictionary comprehension that maps each pixel to its dilation and erosion results.

SciPy

from scipy import ndimage
import numpy as np

image = np.array([[0, 1], [1, 0]])
results = {pixel: (ndimage.binary_[1](image == pixel), ndimage.binary_[2](image == pixel)) for pixel in [0, 1]}
print(results)

Drag options to blanks, or click blank then click option'

Adilation

Berosion

Copening

Dclosing

Attempts:

3 left

💡 Hint

Common Mistakes

Swapping dilation and erosion

Using opening or closing instead

Not using binary_ functions

The dictionary maps each pixel value to its dilation and erosion results using binary_dilation and binary_erosion.

5fill in blank

hard

Fill all three blanks to create a dictionary comprehension that maps each pixel to its closing, erosion, and opening results.

SciPy

from scipy import ndimage
import numpy as np

image = np.array([[1, 0, 1], [0, 1, 0], [1, 0, 1]])
results = {pixel: (ndimage.binary_[1](image == pixel), ndimage.binary_[2](image == pixel), ndimage.binary_[3](image == pixel)) for pixel in [0, 1]}
print(results)

Drag options to blanks, or click blank then click option'

Adilation

Berosion

Copening

Dclosing

Attempts:

3 left

💡 Hint

Common Mistakes

Mixing up opening and closing

Using dilation instead of closing

Not using binary_ functions

The dictionary comprehension maps each pixel to its closing (binary_closing), erosion (binary_erosion), and opening (binary_opening) results.

Practice

1. What does the erosion operation do to a binary image in morphological processing?

easy

A. It inverts the colors of the image.

B. It enlarges the white regions by adding pixels to edges.

C. It shrinks the white regions by removing edge pixels.

D. It blurs the image to reduce noise.

Solution

1. Step 1: Understand erosion effect

   Erosion removes pixels on object boundaries, making white regions smaller.

2. Step 2: Compare with other operations

   Dilation adds pixels, inversion changes colors, blurring smooths image; none match erosion.

3. Final Answer:

   It shrinks the white regions by removing edge pixels. -> Option C

4. Quick Check:

   Erosion = Shrink white regions [OK]

Hint: Erosion shrinks shapes by cutting edges [OK]

Common Mistakes:

• Confusing erosion with dilation
• Thinking erosion adds pixels
• Mixing erosion with color inversion

2. Which of the following is the correct way to import the erosion function from scipy's ndimage module?

easy

A. from scipy.ndimage import erosion

B. from scipy.ndimage import binary_erosion

C. import scipy.ndimage.erosion

D. from scipy import erosion

Solution

1. Step 1: Identify correct function name

   The function for erosion on binary images is named binary_erosion in scipy.ndimage.

2. Step 2: Check import syntax

   Correct import is from scipy.ndimage import binary_erosion. Other options are invalid or incorrect names.

3. Final Answer:

   from scipy.ndimage import binary_erosion -> Option B

4. Quick Check:

   Correct import = binary_erosion [OK]

Hint: Use binary_erosion for erosion in scipy.ndimage [OK]

Common Mistakes:

• Using wrong function name 'erosion'
• Incorrect import syntax
• Trying to import from scipy root

3. Given the following code, what will be the output array after applying dilation?

import numpy as np
from scipy.ndimage import binary_dilation

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

result = binary_dilation(image).astype(int)
print(result)

medium

A. [[0 0 0 0 0]
[0 0 1 0 0]
[0 1 0 0 0]
[0 0 0 0 0]]

B. [[0 0 0 0 0]
[0 1 1 0 0]
[0 1 0 0 0]
[0 0 0 0 0]]

C. [[1 1 1 0 0]
[1 1 1 1 0]
[1 1 1 1 0]
[0 1 1 0 0]]

D. [[0 1 1 0 0]
[1 1 1 1 0]
[1 1 1 0 0]
[0 1 0 0 0]]

Solution

1. Step 1: Understand binary_dilation effect

   Dilation adds pixels to the edges of white regions (1s), expanding them by one pixel in all directions.

2. Step 2: Apply dilation to given image

   Original white pixels at (1,1),(1,2),(2,1). After dilation, neighbors become 1, resulting in the array in [[0 1 1 0 0] [1 1 1 1 0] [1 1 1 0 0] [0 1 0 0 0]].

3. Final Answer:

   [[0 1 1 0 0] [1 1 1 1 0] [1 1 1 0 0] [0 1 0 0 0]] -> Option D

4. Quick Check:

   Dilation = expand edges [OK]

Hint: Dilation grows white pixels by one layer [OK]

Common Mistakes:

• Confusing dilation with erosion
• Not converting boolean to int for print
• Misreading array indices

4. The following code is intended to perform erosion on a binary image, but it raises an error. What is the problem?

import numpy as np
from scipy.ndimage import erosion

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

result = erosion(image)
print(result)

medium

A. The function 'erosion' does not exist in scipy.ndimage; use 'binary_erosion' instead.

B. The input image must be float type, not int.

C. The image array shape is invalid for erosion.

D. The print statement syntax is incorrect.

Solution

1. Step 1: Check function availability

   Scipy.ndimage does not have a function named 'erosion'; the correct function is 'binary_erosion'.

2. Step 2: Correct the import and usage

   Replace 'from scipy.ndimage import erosion' with 'from scipy.ndimage import binary_erosion' and call 'binary_erosion(image)'.

3. Final Answer:

   The function 'erosion' does not exist in scipy.ndimage; use 'binary_erosion' instead. -> Option A

4. Quick Check:

   Use binary_erosion, not erosion [OK]

Hint: Use binary_erosion, not erosion function [OK]

Common Mistakes:

• Trying to import non-existent 'erosion'
• Ignoring error messages
• Assuming all morphological functions have simple names

5. You have a noisy binary image with small white dots scattered outside the main object. Which sequence of morphological operations using scipy.ndimage would best remove these small dots but keep the main shape mostly intact?

hard

A. Apply erosion followed by dilation (opening) to remove small objects.

B. Apply dilation followed by erosion (closing) to fill small holes.

C. Apply only dilation to enlarge all white areas.

D. Apply only erosion to shrink all white areas drastically.

Solution

1. Step 1: Understand noise removal goal

   Small white dots are noise; we want to remove them without changing main shape much.

2. Step 2: Choose correct morphological sequence

   Opening (erosion then dilation) removes small objects but keeps main shape. Closing fills holes, not remove dots.

3. Final Answer:

   Apply erosion followed by dilation (opening) to remove small objects. -> Option A

4. Quick Check:

   Opening = erosion + dilation removes noise [OK]

Hint: Use opening (erosion then dilation) to remove small noise [OK]

Common Mistakes:

• Using closing instead of opening for noise removal
• Applying only dilation or erosion alone
• Confusing noise removal with hole filling