Image properties (shape, dtype, size) in Computer Vision - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ Predict Output

intermediate

What is the shape of the image array?

Given the following code that loads an image as a NumPy array, what is the shape of the array?

Computer Vision

import numpy as np
image = np.zeros((128, 256, 3), dtype=np.uint8)
print(image.shape)

A (128, 256, 3)

B (3, 128, 256)

C (256, 128, 3)

D (128, 256)

💡 Hint

Remember the shape format is (height, width, channels) for color images.

The shape (128, 256, 3) means the image has 128 pixels in height, 256 pixels in width, and 3 color channels (RGB).

❓ Predict Output

intermediate

What is the data type of the image array?

What will be printed by this code snippet?

Computer Vision

import numpy as np
image = np.ones((64, 64), dtype=np.float32)
print(image.dtype)

A float64

B float32

C uint8

D int64

💡 Hint

Check the dtype argument used when creating the array.

The dtype was set to np.float32, so the printed data type is float32.

❓ Metrics

advanced

Calculate the total number of pixels in a grayscale image

If an image has shape (512, 512), how many pixels does it contain?

A 512

B 1024

C 262144

D 1048576

💡 Hint

Multiply the height by the width to get total pixels.

512 * 512 = 262144 pixels in total.

❓ Model Choice

advanced

Choosing the correct image property for model input

You want to feed a color image into a neural network that expects input shape (batch_size, height, width, channels). Which property must you check before feeding the image?

A Data type of the image array

B Size of the image file on disk

C Number of unique colors in the image

D Shape of the image array

💡 Hint

The model expects a specific shape format.

The model input requires the image array shape to match (batch_size, height, width, channels).

🔧 Debug

expert

Why does this code raise an error when accessing image size?

Consider this code snippet: import numpy as np image = np.zeros((100, 100, 3), dtype=np.uint8) print(image.size())

Computer Vision

import numpy as np
image = np.zeros((100, 100, 3), dtype=np.uint8)
print(image.size())

A TypeError: 'numpy.int64' object is not callable

B Prints 30000

C Prints (100, 100, 3)

D SyntaxError due to missing colon

💡 Hint

Check if size is a method or a property.

The 'size' attribute is a property, not a method, so calling size() raises a TypeError.