各向异性扩散修正滤波器的matlab仿真+仿真录像

function [ I ] = pa_diffusion( varargin ) % #1. pa_diffusion(I,dt,kappa) - uses article CG % #2. pa_diffusion(I,dt,kappa,tol) - uses bult in CG with tolerance if (nargin==3) I = varargin{1}; dt = varargin{2}; kappa = varargin{3}; elseif (nargin==4) I = varargin{1}; dt = varargin{2}; kappa = varargin{3}; tol = varargin{4}; else error('pa_diffusion accepts up to 4 input arguments!') end [n,m] = size(I); x = reshape(I',1,n*m); if(nargin==4) x = x'; end [A,b,kappa] = getAb(I,dt,kappa); prevIm = x+2; while(max(abs(x-prevIm))>1) prevIm = x; if (nargin==4) % #1. https://www.mathworks.com/help/matlab/ref/pcg.html matrixMult = @(v) multAx(A,v,n,m)'; x = pcg(matrixMult,b',tol,[],[],[],x); else % #2. article proposal r = b - multAx(A,x,n,m); z = r; prevX = 10^10; while (max(abs(x-prevX))>1 && sum(r~=0)) prevX = x; az = multAx(A,z,n,m); dotr = dot(r,r); alpha = dotr/dot(az,z); x = x + alpha*z; r = r -alpha*az; betta = dot(r,r)/dotr; z = r + betta*z; end end I = reshape(x,m,n)'; [A,b,kappa] = getAb(I,dt,kappa); end end