This is a python implementation of the structural similarity approach to assess image quality. It is based on "Image quality assessment: from error visibility to structural similarity." by Wang et al. and the implementation of scikit-image.
The script takes the paths to the image file and the reference image file.
It optionally plots the computed structural similarity map.
For usage see ./ssim.py -h
The script requires the following python libraries:
- pillow
- NumPy
- SciPy
- matplotlib (optional, for plotting)