Document layout analysis in Computer Vision - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to load an image for document layout analysis.

Computer Vision

from PIL import Image
image = Image.open([1])

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

Aopen('document.jpg')

Bdocument.jpg

C'document.jpg'

DImage.load('document.jpg')

Attempts:

3 left

💡 Hint

Common Mistakes

Forgetting to put quotes around the filename.

Using Image.load instead of Image.open.

The Image.open() function requires the file path as a string, so it must be enclosed in quotes.

2fill in blank

medium

Complete the code to convert the image to grayscale for easier layout analysis.

Computer Vision

gray_image = image.convert([1])

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

A'L'

B'P'

C'CMYK'

D'RGB'

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'RGB' instead of 'L' for grayscale conversion.

Using modes like 'CMYK' which are for printing.

The mode 'L' converts the image to grayscale (luminance), which simplifies layout analysis.

3fill in blank

hard

Fix the error in the code to detect contours using OpenCV for layout blocks.

Computer Vision

import cv2
contours, _ = cv2.findContours(gray_image, cv2.RETR_EXTERNAL, [1])

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

Acv2.RETR_LIST

Bcv2.CHAIN_APPROX_SIMPLE

Ccv2.CHAIN_APPROX_NONE

Dcv2.RETR_TREE

Attempts:

3 left

💡 Hint

Common Mistakes

Using a retrieval mode where an approximation method is expected.

Using CHAIN_APPROX_NONE which is less efficient.

CHAIN_APPROX_SIMPLE compresses horizontal, vertical, and diagonal segments and leaves only their end points, which is efficient for layout detection.

4fill in blank

hard

Fill both blanks to filter contours by area and draw bounding boxes.

Computer Vision

for cnt in contours:
    area = cv2.contourArea(cnt)
    if area [1] 1000:
        x, y, w, h = cv2.boundingRect(cnt)
        cv2.rectangle(image, (x, y), (x + w, y + h), [2], 2)

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

A>

B<

C(0, 255, 0)

D(255, 0, 0)

Attempts:

3 left

💡 Hint

Common Mistakes

Using '<' instead of '>' to filter contours.

Using red color instead of green for rectangles.

We want to keep contours with area greater than 1000 to ignore small noise. The rectangle color is green (0,255,0).

5fill in blank

hard

Fill all three blanks to prepare data for a simple ML model to classify layout blocks.

Computer Vision

features = []
labels = []
for block in layout_blocks:
    features.append([block.[1], block.[2]])
    labels.append(block.[3])

from sklearn.model_selection import train_test_split
X_train, X_test, y_train, y_test = train_test_split(features, labels, test_size=0.2, random_state=42)

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

Awidth

Bheight

Clabel

Dcolor

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'color' as a feature or label which is not typical here.

Mixing up features and labels.

Width and height are common features for layout blocks. The label is the target variable for classification.