MATLAB的图像提取形态学实战之对特殊形状物体的识别确定——圆形目标的识别计算

MATLAB的图像提取形态学实战之对特殊形状物体的识别确定——圆形目标的识别计算

clear all;close all;clc;

RGB01=imread('X233.jpg');
subplot(221);
imshow(RGB01);title('原始图像');

I01=rgb2gray(RGB01);
threshold=graythresh(I01);
bw01=im2bw(I01,threshold);%阈值分割二值化
subplot(222);
imshow(bw01);title('二值图像');

bw02=bwareaopen(bw01,5);%形态学开运算去除像素小于10的目标
subplot(223);
imshow(bw02);title('开运算去除噪声');

sse=strel('disk',2);
bw03=imclose(bw02,sse);%对图像进行填充缝隙和孔洞
%寻找图像目标的边界
[B,L]=bwboundaries(bw03,'noholes');
subplot(224);hold on;
imshow(label2rgb(L,@jet,[.1 .1 .1]));title('检测目标图像');
 for k =1:length(B)
     boundary=B{k};
     plot(boundary(:,2),boundary(:,1),'y','LineWidth',1);
 end

 stats=regionprops(L,'Area','Centroid');
 threshold=0.95;
 
 for k =1:length(B)
     boundary=B{k};
     delta_sq=diff(boundary).^2;
     perimeter=sum(sqrt(sum(delta_sq,2)));
     
     area=stats(k).Area;
     
     metric=4*pi*area/perimeter^2;
     
     metric_str=sprintf('%2.2f',metric);
     if metric>threshold
         centroid=stats(k).Centroid;
         plot(centroid(1),centroid(2),'ko');
     end
     
     text(boundary(1,2)-35,boundary(1,1)+13,metric_str,...
         'Color','y','FontSize',14,'FontWeight','bold');
 end     

运行效果如图：

图片素材不是太好，不过算法从理论上看，没有问题。