Histogram equalization in Computer Vision - Practice Problems & Coding Challenges

Challenge - 5 Problems

🧠 Conceptual

intermediate

What is the main goal of histogram equalization?

Histogram equalization is a technique used in image processing. What does it primarily aim to improve in an image?

A Increase the contrast by spreading out the most frequent intensity values

B Blur the image to reduce noise

C Convert the image to grayscale

D Reduce the image size without losing quality

💡 Hint

Think about how histogram equalization changes the distribution of pixel brightness.

Histogram equalization improves image contrast by redistributing pixel intensities to use the full range of possible values.

❓ Predict Output

intermediate

Output of histogram equalization on a simple grayscale image

Given the following grayscale image pixel values, what will be the pixel values after applying histogram equalization?

Computer Vision

import numpy as np
from skimage import exposure

image = np.array([[50, 50, 80], [80, 100, 100], [150, 150, 200]], dtype=np.uint8)
equalized = exposure.equalize_hist(image)
result = (equalized * 255).astype(np.uint8)
print(result)

A

[[  0   0  85]
 [ 85 170 170]
 [213 213 255]]

B

[[ 50  50  80]
 [ 80 100 100]
 [150 150 200]]

C

[[  0  85  85]
 [170 170 255]
 [255 255 255]]

D

[[  0  42  85]
 [128 170 213]
 [213 255 255]]

💡 Hint

Histogram equalization maps the lowest pixel value to 0 and the highest to 255, spreading others evenly.

The pixel values are rescaled to cover the full 0-255 range, preserving relative order but increasing contrast.

❓ Model Choice

advanced

Choosing the right histogram equalization method for color images

You want to apply histogram equalization to a color image without distorting its colors. Which method is best?

A Apply histogram equalization on the blue channel only

B Convert the image to HSV and apply equalization only on the V channel

C Apply histogram equalization on the grayscale version of the image

D Apply histogram equalization separately on each RGB channel

💡 Hint

Think about which channel controls brightness without changing color.

Equalizing the V (value) channel in HSV space adjusts brightness while preserving hue and saturation, avoiding color distortion.

❓ Hyperparameter

advanced

Effect of clip limit in CLAHE (Contrast Limited Adaptive Histogram Equalization)

In CLAHE, what does increasing the clip limit parameter do to the output image?

A It reduces the image resolution

B It decreases contrast and smooths the image

C It converts the image to grayscale

D It increases contrast enhancement but may cause noise amplification

💡 Hint

Higher clip limit means allowing more contrast stretching locally.

A higher clip limit allows stronger contrast enhancement but can also amplify noise in uniform areas.

❓ Metrics

expert

Evaluating histogram equalization impact using image entropy

After applying histogram equalization to an image, which of the following statements about image entropy is true?

A Entropy becomes zero because the image is normalized

B Entropy decreases because the image becomes simpler

C Entropy usually increases because pixel intensity distribution becomes more uniform

D Entropy remains the same because pixel values are only shifted

💡 Hint

Entropy measures randomness or information content in pixel intensities.

Histogram equalization spreads out pixel intensities, increasing randomness and thus entropy.