32边界的链码表示&MPP算法MATLAB实现

原文：http://3y.uu456.com/bp_3psix2u34h9da6a52j1q_1.html

参考代码：

clc

clear

f = imread('Fig1102(a)(noisy_circular_stroke).tif');

subplot(2,2,1);

imshow(f)

title('原始图像')

%%%平滑图像%%%%

h = fspecial('average',9);

g = imfilter(f,h,'replicate');

subplot(2,2,2)

imshow(g)

title('平滑图像')

%%%%%%%%阈值分割

g = im2bw(g,0.5);

subplot(2,2,3)

imshow(g)

title('阈值分割结果')

B = boundaries(g);

d = cellfun('length',B);

[max_d,k] = max(d);

b = B{1};

[M,N] = size(g); % 注意 : 产生了完整的边界, 边界并没有断开, 拖动一下便可以显现

g = bound2im(b,M,N,min(b(:,1)),min(b(:,2)));

subplot(2,2,4)

imshow(g)

title('边界图像')

[s,su] = bsubsamp(b,50); %对边界进行二次取样

g2 = bound2im(s,M,N,min(s(:,1)),min(s(:,2)));

%subplot(2,3,5)

figure;

imshow(g2)

title('取样结果图')

cn = connectpoly(s(:,1),s(:,2));

g2 = bound2im(cn,M,N,min(cn(:,1)),min(cn(:,2)));

%subplot(2,3,6)

figure;

imshow(g2)

title('取样结果连接图')

c = fchcode(su)

2.使用函数 minperpoly 实现 MPP 算法，得到边界的多边形近似。 参考代码：

clc

clear

B = imread('Fig1107(a)(mapleleaf).tif');figure;subplot(121); imshow(B)

title('原始图像')

b = boundaries(B,4,'cw'); %4 连接边界

b = b{1};

[M,N] = size(B);

xmin = min(b(:,1));

ymin = min(b(:,2));

bim = bound2im(b,M,N,xmin,ymin);

subplot(122);

imshow(bim);title('边界图像')

[x,y] = minperpoly(B,2); %使用大小为 2 的方形单元得到的 MPP b2 = connectpoly(x,y);

B2 = bound2im(b2,M,N,xmin,ymin);

figure;subplot(131);

imshow(B2);title('2方形单元')

[x,y] = minperpoly(B,4); %使用大小为 4 的方形单元得到的 MPP b2 = connectpoly(x,y);

B2 = bound2im(b2,M,N,xmin,ymin);

subplot(132);

imshow(B2);title('4方形单元')

[x,y] = minperpoly(B,16); %使用大小为 16 的方形单元得到的 MPP b2 = connectpoly(x,y);

B2 = bound2im(b2,M,N,xmin,ymin);

subplot(133);

imshow(B2);title('16方形单元')