Python CV ecosystem (OpenCV, PIL, torchvision) in Computer Vision - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to read an image using OpenCV.

Computer Vision

import cv2
image = cv2.[1]('image.jpg')

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

Aimread

Bread

Copen

Dload

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'cv2.read' instead of 'cv2.imread'.

Trying to use 'open' which is a Python built-in, not OpenCV.

The function cv2.imread is used to read an image file into an array.

2fill in blank

medium

Complete the code to convert a PIL image to a tensor using torchvision.

Computer Vision

from torchvision import transforms
transform = transforms.ToTensor()
tensor_image = transform([1])

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

Anumpy_array

Bimage_tensor

Ccv_image

Dpil_image

Attempts:

3 left

💡 Hint

Common Mistakes

Passing a numpy array instead of a PIL image.

Passing an OpenCV image which is a numpy array with BGR channels.

The ToTensor() transform converts a PIL Image or numpy.ndarray to a tensor. Here, the input must be a PIL image.

3fill in blank

hard

Fix the error in the code to convert an OpenCV image to RGB format.

Computer Vision

import cv2
image_bgr = cv2.imread('image.jpg')
image_rgb = cv2.[1](image_bgr, cv2.COLOR_BGR2RGB)

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

AconvertColor

BcvtColor

CchangeColor

DtransformColor

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'convertColor' which does not exist in OpenCV.

Using 'transformColor' which is invalid.

The correct OpenCV function to convert color spaces is cvtColor.

4fill in blank

hard

Fill both blanks to create a PIL image from a numpy array and save it.

Computer Vision

from PIL import Image
import numpy as np
array = np.zeros((100, 100, 3), dtype=np.uint8)
img = Image.[1](array)
img.[2]('output.png')

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

Afromarray

Bsave

Cshow

Dopen

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'Image.open' instead of 'Image.fromarray'.

Using 'img.show()' instead of 'img.save()' to save the image.

Image.fromarray creates a PIL image from a numpy array. save saves the image to a file.

5fill in blank

hard

Fill all three blanks to normalize a tensor image using torchvision transforms.

Computer Vision

from torchvision import transforms
normalize = transforms.Normalize(mean=[[1]], std=[[2]])
transform = transforms.Compose([transforms.ToTensor(), normalize])
# Apply transform to PIL image
normalized_tensor = transform(pil_image)
print(normalized_tensor.mean().item() [3] 0)

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

A0.485, 0.456, 0.406

B0.229, 0.224, 0.225

C>

D<

Attempts:

3 left

💡 Hint

Common Mistakes

Swapping mean and std values.

Using '<' instead of '>' in the print statement.

The standard ImageNet mean and std values are used for normalization. After normalization, the mean of the tensor should be greater than 0.