This is my implementation of the SSEQ index. I wasn't able to find a fully implemented Python version of this index, so I decided to use Aca4peop's code as a starting point and then add my own modifications.
The full details of SSEQ can be found in the paper: No-reference image quality assessment based on spatial and spectral entropies (Liu et al.). The original MATLAB implementation is here.
Vectorized implementation of:
- Patch spatial entropy
- DCT for spectral entropy (more info here)
Every dataset was split into a training and a test set. I used the training sets with K-fold cross-validation to get the best parameters for each SVR model. The following are the results on each test set:
| Dataset | LCC | SROCC |
|---|---|---|
| csiq | 0.8493 | 0.7913 |
| kadid10k | 0.6075 | 0.5716 |
| koniq10k | 0.5745 | 0.5573 |
| liveiqa | 0.8726 | 0.8713 |
| tid2013 | 0.7892 | 0.7204 |