Clean Up Drawing Scans with Python

This Python script serves as a tool to clean up scanned pen and ink illustrations. It applies a series of image processing techniques to enhance the quality of the input image and remove imperfections.

Requirements

To run the script, you need to have the following dependencies installed:

  • OpenCV (cv2): A popular computer vision library.
  • NumPy (numpy): A powerful library for numerical computing.
  • argparse (argparse): A module for command-line argument parsing.
  • logging (logging): A module for logging messages during execution.

Usage

To use the script, execute it with the following command-line arguments:

python script_name.py input_image output_image [--debug]
  • input_image: The path to the input pen and ink illustration image that requires cleaning.
  • output_image: The path where the cleaned image will be saved.
  • --debug: (Optional) Enable debug mode to receive more detailed logging information.

Image Cleaning Steps

  1. Image Loading: The script loads the input image in grayscale mode using OpenCV.

Test Image

  1. Gaussian Blur: A Gaussian blur is applied to the image to smoothen it and reduce noise.

Gaussian Blur

  1. Adaptive Thresholding: The script uses adaptive thresholding to dynamically threshold the blurred image. This technique helps to separate the foreground (pen and ink lines) from the background.

Otsu Threshold

  1. Morphological Operations:
    • Opening: Morphological opening is performed to remove small dots and artifacts from the image.
    • Closing: A closing operation is applied to close small gaps in the ink lines.

Opened

Closed

  1. Image Inversion: The cleaned image is inverted to preserve the pen and ink lines while discarding the background.

  2. Combining Images: The script performs a bitwise OR operation between the inverted cleaned image and the original image. This combines the cleaned ink lines with the original content, resulting in a cleaned and enhanced image.

Result

Example

python image_cleaning_script.py input.png output_cleaned.png --debug

This command will load the input.png image, clean it using the defined techniques, and save the resulting cleaned image as output_cleaned.png. Debugging messages will be displayed during the process.

Note: Ensure that you have the necessary permissions to read the input image and write the output cleaned image in the specified locations.

Script