Find defects based on differnces between two images

Detect differences between two images to find defects

Code environment:

Python 3.11.2

See requirements.txt for used libraries

Assumptions:

• Grayscale images with 3 graded intensities as input
• Images size around 300-500 pixels in each direction
• Uses translation registration: assumes no shearing, reflection, scaling or rotation between images
• Should be able to deal with both small and large defects - avoids strong filtering
• Favors false positive over false negatives