cbi_toolbox.reconstruct¶
Submodules
Module contents
The reconstruct package provides reconstruction algorithms, as well as preprocessing tools and performance scores.
-
cbi_toolbox.reconstruct.mse(ref, target)[source]¶ Computes the Mean Squared Error between two arrays
- Parameters
ref (numpy.ndarray) – Reference array.
target (numpy.ndarray) – Target array.
- Returns
The MSE.
- Return type
float
-
cbi_toolbox.reconstruct.normalize(image, mode='std')[source]¶ Normalize an image according to the given criterion.
- Parameters
image (numpy.ndarray) – Image to normalize, will be modified.
mode (str, optional) – Type of normalization to use, by default ‘std’. Allowed: [‘std’, ‘max’, ‘sum’]
- Returns
The normalized image (same as input).
- Return type
array
- Raises
ValueError – For unknown mode.
-
cbi_toolbox.reconstruct.psnr(ref, target, norm=None, limit=None)[source]¶ Computes the Peak Signal-to-Noise Ratio: PSNR = 10 log( limit ^ 2 / MSE(ref, target) )
- Parameters
ref (numpy.ndarray) – The ground-truth reference array.
target (numpy.ndarray) – The reconstructed array.
norm (str) – Normalize the images before computing snr, default is None.
limit (float, optional) – The maximum pixel value used for PSNR computation, default is None (max(ref)).
- Returns
The PSNR.
- Return type
float