bill2462 / spongebob

Calculate material porosity based on image processing using openCV

Sponge Bob

Calculate porosity by processing microscopic images using openCV.

Instalation

python3 setup.py bdist_wheel
cd dist
pip3 install --upgrade spongeBob-0.1-py3-none-any.whl

Usage

spongeBob [-h] --output-file OUTPUT_FILE dataInput

As a dataInput program takes:

• Path to a directory containing images.
• Path to a image file.

Program can save results into a CSV file. Option --output-file takes a path to this file.

Example:

python3 -m spongeBob /home/patrik/porositySamples --output-file=/home/patrik/porosity.csv

Example output:

6 images detected in '/home/patrik/porosity'
Loading file /home/patrik/porosity/8bars - center.jpg ...
Loading file /home/patrik/porosity/8bars - edge.jpg ...
Loading file /home/pqtrik/porosity/12bars - center.jpg ...
Loading file /home/patrik/porosity/12bars - edge.jpg ...
Loading file /home/patrik/porosity/15bars - center.jpg ...
Loading file /home/patrik/porosity/15bars - edge.jpg ...
Processing image '/home/patrik/porosity/8bars - center.jpg' ...
Porosity: 39.35546875%
Processing image '/home/patrik/porosity/8bars - edge.jpg' ...
Porosity: 48.59822591145833%
Processing image '/home/patrik/porosity/12bars - center.jpg' ...
Porosity: 45.508544921875%
Processing image '/home/patrik/porosity/12bars - edge.jpg' ...
Porosity: 40.063313802083336%
Processing image '/home/patrik/porosity/15bars - center.jpg' ...
Porosity: 38.288167317708336%
Processing image '/home/patrik/porosity/15bars - edge.jpg' ...
Porosity: 25.493896484375%
Exporting results to /home/patrik/output.csv ...

Example csv output file:

filename	porosity
/home/patrik/porosity/8bars - center.jpg	39.35546875
/home/patrik/porosity/8bars - edge.jpg	48.59822591145833
/home/patrik/porosity/12bars - center.jpg	45.508544921875
/home/patrik/porosity/12bars - edge.jpg	40.063313802083336
/home/patrik/porosity/15bars - center.jpg	38.288167317708336
/home/patrik/porosity/15bars - edge.jpg	25.493896484375

How does it work?

Let's consider the sponge image.

First the image is converted to a grayscale.

Then the avarage brightness of the image is calculated and all pixels with the value below avarage are changed to 0 and all other pixels are changed to 255.

Then the program calculates the porosity by calculating what percentage of all pixels in the processed image is taken by the dark pixels (pixels with value 0).