Checkbox Detection and Classification

This Python script detects and classifies checkboxes in a given image. It identifies checkboxes in forms or documents and determines if they are ticked or unticked. The script uses OpenCV for image processing and numpy for numerical operations. The results are displayed visually and printed in a readable format.

It also contains the jupyter notebook for executing the files in a folder and plot the results using matplotlib

Requirements

• Python 3.x
• OpenCV (cv2)
• NumPy (numpy)
• Matplotlib (matplotlib)

Installation

To install the required libraries, you can use pip:

pip install opencv-python-headless numpy matplotlib

Usage

Run the script from the command line, providing the path to the image file as an argument:

python script.py path/to/your/image.png

Description

Functions

1. detect_checkboxes(image, line_min_width=15, line_max_width=15)

   • Converts the image to grayscale.
   • Binarizes the image to highlight potential checkboxes.
   • Uses morphological operations to detect horizontal and vertical lines.
   • Combines the results to identify checkbox-like structures.
   • Finds connected components in the processed image.
   • Returns the statistics and labels of the detected components.

2. classify_checkboxes(image, stats, min_width=10, max_width=50, tick_threshold=0.1)

   • Filters components based on size and aspect ratio to identify checkboxes.
   • Extracts the region of interest (ROI) for each checkbox.
   • Preprocesses the ROI by converting to grayscale, applying Gaussian blur, and binarizing.
   • Counts non-white pixels in the binarized ROI to determine if the checkbox is ticked.
   • Classifies and labels the checkbox as 'Ticked' or 'Unticked'.
   • Draws rectangles and labels on the image for visualization.
   • Returns a list of classified checkboxes.

3. process_image(image_path)

   • Reads the image from the provided path.
   • Calls detect_checkboxes and classify_checkboxes to process the image.
   • Displays the processed image using Matplotlib.
   • Returns the list of classified checkboxes.

Command-Line Interface

The script uses the sys module to handle command-line arguments. The image path is provided as an argument, and the script processes the image and prints the results.

Example Output

When you run the script, it will display the processed image and print the results in the following format:

path/to/your/image.png: [(694, 109, 10, 12, 'Unticked'), (146, 813, 25, 26, 'Ticked'), ...]
(694, 109, 10, 12, 'Unticked')
(146, 813, 25, 26, 'Ticked')
...

Each tuple in the list represents a detected checkbox with its coordinates, dimensions, and classification ('Ticked' or 'Unticked').

Example

To process an image named test_image1.jpg:

python script.py test_image1.jpg

The script will output:

test_image1.jpg: [(694, 109, 10, 12, 'Unticked'), (146, 813, 25, 26, 'Ticked'), (352, 813, 25, 25, 'Ticked'), (615, 813, 25, 25, 'Ticked'), (905, 813, 25, 25, 'Ticked'), (1249, 813, 25, 25, 'Ticked')]
(694, 109, 10, 12, 'Unticked')
(146, 813, 25, 26, 'Ticked')
(352, 813, 25, 25, 'Ticked')
(615, 813, 25, 25, 'Ticked')
(905, 813, 25, 25, 'Ticked')
(1249, 813, 25, 25, 'Ticked')

And display the image with highlighted checkboxes.

---

Credits

Special thanks to:

This project was inspired by the work of Sreekiranar.