demo chenvese.m

% demo for chenvese function
% Copyright (c) 2009, 
% Yue Wu @ ECE Department, Tufts University
% All Rights Reserved  
% http://sites.google.com/site/rexstribeofimageprocessing/
%%
%-- Chan & Vese method on gray and color image
%   Find contours of objects
close all
clear all

I = imread('brain.jpg');
% Customerlized Mask
m = zeros(size(I,1),size(I,2));
m(20:120,20:120) = 1;
seg = chenvese(I,m,500,0.1,'chan'); % ability on gray image
% Built-in Mask
seg = chenvese(I,'medium',400,0.02,'chan'); % ability on gray image
%-- End 

%%
%-- Chan & Vese method on RGB/vector image
%   Strong ability to resist noise
close all
clear all
P = imread('anti-mass.jpg');
% Imnoise the original input
I = P;
I(:,:,1) = imnoise(I(:,:,1),'speckle');
I(:,:,2) = imnoise(I(:,:,2),'salt & pepper',0.8);
figure(),subplot(1,2,1),imshow(P),title('original image');
subplot(1,2,2),imshow(I),title('original image with two components adding noise')

% Normal Chan & Vese cannot work
seg = chenvese(I,'large',300,0.02,'chan'); 

% Chan & Vese for vector image works here
seg = chenvese(I,'large',300,0.02,'vector');
% Using built-in mask = 'whole' leads faster and better segmentation
seg = chenvese(I,'whole',800,0.02,'vector');

%-- End 

%%
%-- Chan & Vese method for multiphase (Here I use only two phases)
% sigle phase can only distinguish two levels of objects 
% (background and foreground)in image
% two phases can distingush four levels of objects
% n phases can distinguish 2^n levels of objects

%------------------------------------------
% example on systhesis image
close all
clear all

I = imread('4colors.jpg');
seg = chenvese(I,'whole',100,0.1,'multiphase'); 
%-------------------------------------------
% example on real image
close all
clear all

I = imread('flowers.jpg');
seg = chenvese(I,'whole',400,0.2,'multiphase'); 
%-- End