Why image processing transforms visual data in SciPy - Challenge Your Understanding

Challenge - 5 Problems

❓ Predict Output

intermediate

Effect of Gaussian Blur on an Image Array

What is the output shape and type after applying Gaussian blur to a 5x5 image array using scipy?

SciPy

import numpy as np
from scipy.ndimage import gaussian_filter

image = np.arange(25).reshape(5,5)
blurred = gaussian_filter(image, sigma=1)
print(type(blurred), blurred.shape)

A <class 'list'> (5, 5)

B <class 'list'> (25,)

C <class 'numpy.ndarray'> (25,)

D <class 'numpy.ndarray'> (5, 5)

💡 Hint

Gaussian blur keeps the image shape but smooths pixel values.

The gaussian_filter returns a numpy array of the same shape as input, smoothing the image but not changing its dimensions or type.

❓ data_output

intermediate

Histogram Equalization Effect on Pixel Intensity Distribution

After applying histogram equalization to a grayscale image array, what is the sum of all pixel intensities?

SciPy

import numpy as np
from skimage import exposure

image = np.array([[50, 80, 90], [30, 60, 70], [20, 40, 100]], dtype=np.uint8)
equalized = exposure.equalize_hist(image)
sum_pixels = equalized.sum()
print(round(sum_pixels, 2))

A 1.5

B 4.5

C 3.0

D 6.0

💡 Hint

Histogram equalization redistributes intensities but total sum depends on normalization.

equalize_hist normalizes pixel values between 0 and 1, so sum is roughly number of pixels times average intensity (~3.0).

❓ visualization

advanced

Visualizing Edge Detection on a Sample Image

Which option shows the correct edge detection output using Sobel filter on a 3x3 image array?

SciPy

import numpy as np
from scipy import ndimage

image = np.array([[10, 10, 10], [10, 50, 10], [10, 10, 10]])
edge_sobel = ndimage.sobel(image)
print(edge_sobel)

A

[[ 0 40  0]
 [40  0 40]
 [ 0 40  0]]

B

[[ 0  0  0]
 [ 0  0  0]
 [ 0  0  0]]

C

[[40 40 40]
 [40  0 40]
 [40 40 40]]

D

[[10 10 10]
 [10 50 10]
 [10 10 10]]

💡 Hint

Sobel filter highlights edges by calculating gradient magnitude.

The Sobel filter detects edges by computing gradients; the center pixel with higher intensity creates edges around it with value 40.

🧠 Conceptual

advanced

Why Transform Visual Data in Image Processing?

Which reason best explains why image processing transforms visual data?

A To convert images into text documents automatically

B To enhance features and remove noise for better analysis

C To increase the file size of images for storage

D To make images invisible to the human eye

💡 Hint

Think about the purpose of improving image quality.

Image processing transforms images to highlight important features and reduce noise, making analysis easier and more accurate.

🔧 Debug

expert

Identify the Error in Image Thresholding Code

What error does the following code raise when thresholding an image array? import numpy as np image = np.array([[100, 150], [200, 250]]) threshold = 180 binary = image > threshold result = binary.astype(np.int) print(result)

SciPy

import numpy as np
image = np.array([[100, 150], [200, 250]])
threshold = 180
binary = image > threshold
result = binary.astype(int)
print(result)

A AttributeError: module 'numpy' has no attribute 'int'

B TypeError: unsupported operand type(s) for >: 'list' and 'int'

C SyntaxError: invalid syntax in line with astype

D No error, prints binary array correctly

💡 Hint

Check if np.int is deprecated or removed in recent numpy versions.

np.int is deprecated and removed in recent numpy versions; use int or np.int32 instead.