Transfer Functions

Any transfer function is a subtype of the abstract TransferFunction

TransferFunctions.TransferFunction — Type

TransferFunction

Super type for all transfer functions

A transfer function in microscopy is an object specifying the response in the image plane to the light coming from the object plane. There are several characteristics of a transfer functions which can influence the character of the response. Most commonly we are concerned with its spatial variance/invariance, i.e. whether the response is dependent of the objects position in the object plane, and linearity (which is almost always satisfied).

source

A transfer function can be either linear or non-linear and shift invariant or shift variant.

TransferFunctions.LinearTransferFunction — Type

LinearTransferFunction{N} <: TransferFunction{N}

An abstract type for all linear transfer functions.

source

TransferFunctions.NonLinearTransferFunction — Type

NonLinearTransferFunction{N} <: TransferFunction{N}

A supertype for all non-linear transfer functions.

source

Any fully prescribed transfer function can be used to perform the forward pass (i.e. optical transfer)

TransferFunctions.transfer — Method

transfer(A::SpatialMatrix, t::TransferFunction)

The forward pass of the transfer function operator t on the image A.

See also restore

source

The reverse is much harder and we need estimation and inverse modeling methods for it

TransferFunctions.restore — Method

restore(A::SpatialMatrix, t::TransferFunction)

The reverse pass of the transfer function operator t on the image A.

It is always inherently an estimation because A is always noisy and t is often not well conditioned (especially in the non-linear case).