Reading images (cv2.imread) in Computer Vision - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ Predict Output

intermediate

What is the shape of the image read by cv2.imread?

Given the code below, what will be the shape of the variable img if the image is a 640x480 pixel color image?

Computer Vision

import cv2
img = cv2.imread('sample.jpg')
print(img.shape)

A (640, 480)

B (640, 480, 3)

C (480, 640)

D (480, 640, 3)

💡 Hint

Remember that OpenCV reads images as height x width x channels.

OpenCV loads images as arrays with shape (height, width, channels). For a 640x480 image, height is 480 and width is 640. Color images have 3 channels (BGR).

🧠 Conceptual

intermediate

What does cv2.imread return if the image file does not exist?

If you try to read an image file that does not exist using cv2.imread('missing.jpg'), what will be the value of the returned variable?

A None

B An empty numpy array with shape (0,0,3)

C Raises a FileNotFoundError

D An array filled with zeros of shape (1,1,3)

💡 Hint

Check what OpenCV returns when it cannot find the file.

cv2.imread returns None if the image file cannot be found or read.

❓ Model Choice

advanced

Which flag reads an image in grayscale using cv2.imread?

You want to read an image as grayscale using cv2.imread. Which flag should you use?

A cv2.IMREAD_GRAYSCALE

B cv2.IMREAD_COLOR

C cv2.IMREAD_UNCHANGED

D cv2.IMREAD_ANYDEPTH

💡 Hint

Grayscale means one channel, black and white image.

cv2.IMREAD_GRAYSCALE reads the image as a single channel grayscale image.

🔧 Debug

advanced

Why does this code raise an AttributeError?

Consider the code below:

import cv2
img = cv2.imread('image.jpg')
print(img.size())

Why does this code raise an AttributeError?

Computer Vision

import cv2
img = cv2.imread('image.jpg')
print(img.size())

A cv2 module does not have imread function

B cv2.imread returns None, so None has no size() method

C 'size' is an attribute, not a method, so calling img.size() causes AttributeError

D The image file is corrupted, causing AttributeError

💡 Hint

Check if size is a method or attribute of numpy arrays.

The 'size' property of numpy arrays is an attribute, not a function. Calling img.size() tries to call a method that does not exist.

❓ Metrics

expert

What is the number of channels in an image read with cv2.IMREAD_UNCHANGED?

If you read a PNG image with transparency using cv2.imread('image.png', cv2.IMREAD_UNCHANGED), how many channels will the resulting image array have?

A 3 channels (Blue, Green, Red)

B 4 channels (Blue, Green, Red, Alpha)

C 1 channel (Grayscale)

D Depends on the image, could be 1 or 3

💡 Hint

IMREAD_UNCHANGED preserves the alpha channel if present.

Using cv2.IMREAD_UNCHANGED reads all channels including alpha, so PNG with transparency has 4 channels.