Image as numerical data (pixels, channels) in Computer Vision - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to load an image as a numerical array using PIL.

Computer Vision

from PIL import Image
import numpy as np
image = Image.open('photo.jpg')
numpy_array = np.array([1])

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

Aimage

BImage

Cnp

Dphoto.jpg

Attempts:

3 left

💡 Hint

Common Mistakes

Passing the module name instead of the image object.

Passing the filename string instead of the image object.

The np.array() function converts the PIL image object image into a numerical array representing pixels.

2fill in blank

medium

Complete the code to get the shape of the image array (height, width, channels).

Computer Vision

height, width, channels = numpy_array[1]

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

A.size

B.shape

C.dtype

D.ndim

Attempts:

3 left

💡 Hint

Common Mistakes

Using .size which gives total number of elements.

Using .dtype which gives data type.

The .shape attribute gives the dimensions of the numpy array as (height, width, channels).

3fill in blank

hard

Fix the error in the code to convert the image to grayscale by averaging color channels.

Computer Vision

grayscale = numpy_array.mean(axis=[1])

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

A2

B1

C0

D-1

Attempts:

3 left

💡 Hint

Common Mistakes

Averaging over axis 0 or 1 which are height or width.

Using negative axis incorrectly.

The color channels are the last dimension (axis 2), so averaging over axis 2 converts the image to grayscale.

4fill in blank

hard

Fill both blanks to create a dictionary mapping pixel coordinates to RGB tuples for pixels with red value > 100.

Computer Vision

pixel_dict = [1]( ((x, y), tuple(numpy_array[x, y])) for x in range(numpy_array.shape[0]) for y in range(numpy_array.shape[1]) if numpy_array[x, y, 0] [2] 100 )

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

Adict

Blist

C>

D<

Attempts:

3 left

💡 Hint

Common Mistakes

Using list instead of dict.

Using < instead of > for the red value condition.

Use dict to create a dictionary. The red channel is index 0, so filter pixels where red > 100.

5fill in blank

hard

Fill all three blanks to normalize pixel values to range 0-1 and get the shape of the normalized array.

Computer Vision

normalized = numpy_array[1] / 255.0
shape = normalized[2]
channels = shape[3]

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

A.astype(float)

B.shape

C[2]

D.astype(int)

Attempts:

3 left

💡 Hint

Common Mistakes

Dividing integers without converting to float.

Using .astype(int) which loses decimal precision.

Accessing wrong index for channels.

Convert to float before division with .astype(float). Use .shape to get dimensions. The channels are at index 2.