基于边缘和GO分布的水平集图像分割

function unew=level_iter(img,u,step,g,alf,beta,gama) %set boundry [nrow,ncol]=size(u); u([1 nrow],[1 ncol]) = u([3 nrow-2],[3 ncol-2]); u([1 nrow],2:end-1) = u([3 nrow-2],2:end-1); u(2:end-1,[1 ncol]) = u(2:end-1,[3 ncol-2]); %****gamma division*** e=1.0; %clculate delta delta=(e/pi)./(e^2+u.^2); % H function H=0.5*(1+(2/pi)*atan(u./e)); %gradient value [ux,uy]=gradient(u); grad=sqrt(ux.^2+uy.^2+1e-10); %curvature Nx=ux./grad; Ny=uy./grad; [nxx,junk]=gradient(Nx); [junk,nyy]=gradient(Ny); K=nxx+nyy; %compute the edge motivation Nxg=g.*Nx; Nyg=g.*Ny; [nxxg,junk]=gradient(Nxg); [junk,nyyg]=gradient(Nyg); gdiv=nxxg+nyyg; %distance adjustment Li=0.04*(4*del2(u)-K); %parameter b=length(find(u>=0)); f=length(find(u<0)); wf=f/(f+b); wb=b/(f+b); %gamma [Pt,Pb]=gamma_dis(img,H); %G0 [pt,pb]=g0_dis(img,H); area=log(wf)+log(pt)-log(wb)-log(Pb); unew=u+step*(alf*delta.*area+beta*delta.*(gdiv+gama*g)+Li);