LION.operators¶

LION operators.

The generic ones are here; each kind of operator has its own folder, with its parameters, operator and reconstructions: LION.operators.ct (computed tomography) and LION.operators.pcm (photocurrent mapping).

Modules

`CompositeOp`(wavelet, phi[, device])	The composite operator A = Phi Psi^-1 and its adjoint: the measurement operator `phi` (e.g. photocurrent mapping's, `LION.operators.pcm.PhotocurrentMapOp`) applied to the image of wavelet coefficients (`wavelet`, a `Wavelet2D`), for sparse recovery in the wavelet domain (Koutsourakis, Thompson and Blakesley, "Toward Megapixel Resolution Compressed Sensing Current Mapping of Photovoltaic Devices Using Digital Light Processing", Solar RRL 5(11):2100467, 2021, doi:10.1002/solr.202100467).
`DebiasOp`(op, y, w[, support_tol])	The forward operator `operator` restricted to the support of `w` (an l1 minimiser; its entries above `support_tol`), for debiasing it by least squares on that support (Koutsourakis, Thompson and Blakesley, "Toward Megapixel Resolution Compressed Sensing Current Mapping of Photovoltaic Devices Using Digital Light Processing", Solar RRL 5(11):2100467, 2021, doi:10.1002/solr.202100467.) `y`: the measurements (M,).
`Operator`([device])	The base class of LION's operators: a linear map, as a torch module, with its adjoint and the shapes it maps between.
`Wavelet2D`(shape[, wavelet_name, level, ...])	The 2D orthogonal wavelet transform (PyWavelets) of images of `shape` (H, W), as an operator: images to a flat vector of coefficients.
`ct`	Computed tomography: the geometry (the CT operator's parameters), the projection operator, noise and HU conversions, and CT reconstructions.
`pcm`	Photocurrent mapping: the photocurrent mapping operator, its subsampling, and the Walsh-Hadamard transform it uses.