Homec4science
Diffusion Motion correction paper (cluster_pipeline) Browse

R6289/code/external/InvPbLib/LinOp.m
3e35ee179701cluster_pipeline

Last Change
Raw File

LinOp.m
View Options

classdef LinOp < handle
%% LinOP : Linear Operator generic class
% Matlab Linear Operator Library
% The LinOp meta class implement generic methods for all linear
% operators $\mathbf{H}$: $\mathcal{X}\rightarrow\mathcal{Y}$
%
% $$ \mathbf{y} = \mathbf{H} \mathbf{x} $$
%
%% Properties
% * |name| - name of the linear operator $\mathbf{H}$
% * |sizein| - dimension of the right hand side vector space $\mathcal{X}$
% * |sizeout| - dimension of the left hand side vector space $\mathcal{Y}$
% * |isinvertible| - true if the operator is invertible
% * |iscomplex| - true is the operator is complex
% * |norm| - norm of the operator (if known, otherwise -1)
%
%% Methods
% * |apply| - apply the operator $\mathbf{H}$
% * |adjoint| - apply the adjoint $\mathbf{H}^{\mathrm{*}}$ defined
% such $<\mathbf{H} \mathbf{x} . \mathbf{y} > = <\mathbf{x} .
% \mathbf{H}^{\mathrm{*}} \mathbf{y} >$
% * |HtH| - apply the HtH matrix: the operator followed by its adjoint
% $\mathbf{H}^{\mathrm{*}} \cdot \mathbf{H}$
% * |HHt| - apply the HHt matrix: the adjoint operator followed by its
% $ \mathbf{H} \cdot \mathbf{H}^{\mathrm{*}} $
% * |inverse| - apply the Inverse $\mathbf{H}^{\mathrm{-1}}$ of the
% operator if it exist
% * |adjointInverse| - apply the adjoint of the Inverse $\mathbf{H}^{\mathrm{-*}}$ of the
% operator if it exist
%
% Copyright (C) 2015 F. Soulez ferreol.soulez@epfl.ch
%
% This program is free software: you can redistribute it and/or modify
% it under the terms of the GNU General Public License as published by
% the Free Software Foundation, either version 3 of the License, or
% (at your option) any later version.
%
% This program is distributed in the hope that it will be useful,
% but WITHOUT ANY WARRANTY; without even the implied warranty of
% MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
% GNU General Public License for more details.
%
% You should have received a copy of the GNU General Public License
% along with this program. If not, see <http://www.gnu.org/licenses/>.
properties (SetAccess = protected,GetAccess = public)
name = 'none' % name of the linear operator
sizein; % dimension of the right hand side vector space
sizeout; % dimension of the left hand side vector space
isinvertible = false; % true if the operator is invertible
iscomplex = false; % true is the operator is complex
norm=-1; % norm of the operator
end
methods
function apply(~,~) % apply the operator
error('Operator not implemented');
end
function adjoint(~,~) % apply the adjoint
error('adjoint not implemented');
end
function y = HtH(this,x) % apply the HtH matrix
y = this.adjoint(this.apply(x));
end
function y = HHt(this,x) % apply the HHt matrix
y = this.apply(this.adjoint(x));
end
function y = HtWH(this,x,W) % apply the HtH matrix
if (isscalar(W) && isreal(W))
y = W.*this.HtH(x);
else
assert(isa(W,'LinOp'),'W must be a LinOp');
y = this.adjoint(W.apply(this.apply(x)));
end
end
function inverse(this,~) % apply the inverse
if this.isinvertible
error('inverse not implemented');
else
error('Operator not invertible');
end
end
function adjointInverse(this,~) % apply the inverse
if this.isinvertible
error('adjointInverse not implemented');
else
error('Operator not invertible');
end
end
function this = transpose(this) % Overloading for transpose
if this.iscomplex
warning('Warning: Do you mean adjoint? For LinOp object transpose() is an alias of adjoint method');
end
this = Adjoint(this);
end
function this = ctranspose(this) % Overloading for ctranspose
this = Adjoint(this);
end
function y = mtimes(this,x)% Overloading for *
if isa(x,'LinOp')
y = MulLinOp(this, x);
else
y = this.apply(x);
end
end
function y = plus(this,x)% Overloading for +
assert(isa(x,'LinOp'),'addition of LinOp is only define with other LinOp');
y = SumLinOp({this,x});
end
function this = setNorm(this, x)
this.norm = x;
end
end
methods (Static, Access=protected)
function checkSize(x, sz)
assert( isequal(size(x),sz), 'x does not have the correct size, [%s]', num2str(sz));
end
end
end
c4science · Help