24小时热门版块排行榜

返回列表

汤和加菲

银虫 (小有名气)

应助: 0 (幼儿园)
金币: 291.5
红花: 1
帖子: 85
在线: 25.1小时
虫号: 3080147
注册: 2014-03-22
专业: 动力学与控制

[求助] 下面的Canny边缘提取代码错在哪里？

下面的程序错在哪里，为什么提取出的边缘图像全黑，而如果删掉高斯滤波代码，反而能够出来结果？

I=imread('F:\rose.jpg');%读入原始图片
figure, imshow(I),title('原图');
Y=rgb2gray(I); %把彩色图片转化成灰度图片，256级
figure, imshow(Y),title('灰度图像');
I1=double(Y);
[m,n]=size(I1);
%（1）用高斯滤波器平滑图像
sigma=2;N=5;%定义高斯参数
[d,f]=meshgrid([-(N-1)/2

N-1)/2],[-(N-1)/2

N-1)/2]);
g=exp(-((d.*d+f.*f)/(2*sigma.*sigma)));
S=conv2(Y,g,'same');%二维卷积得到高斯滤波后的图象
I1=mat2gray(S);%归一化，使矩阵中每个元素的值都在0-1之间
% gaussian=[1/79 2/79 3/79 2/79 1/79; 2/79 4/79 6/79 4/79 2/79; 3/79 6/79 7/79 6/79 3/79; 2/79 4/79 6/79 4/79 2/79; 1/79 2/79 3/79 2/79 1/79];
%  I1 =conv2(I1,gaussian);
imshow(I1);title('高斯滤波结果');
%（2）用一阶偏导的有限差分来计算梯度的幅值和方向；
IP=zeros(m,n);
IQ=zeros(m,n);
f=zeros(m,n);
a=zeros(m,n);
for x=1:m-1
for y=1:n-1
IP(x,y)=(I1(x,y+1)-I1(x,y)+I1(x+1,y+1)-I1(x+1,y))/2;
IQ(x,y)=(I1(x,y)-I1(x+1,y)+I1(x,y+1)-I1(x+1,y+1))/2;
f(x,y)=sqrt(IP(x,y)^2+IQ(x,y)^2);%梯度幅值矩阵
a(x,y)=0;
if(IP(x,y)<0)
      a(x,y)=atan(IQ(x,y)/IP(x,y))+pi/2;
else
      a(x,y)=atan(IQ(x,y)/IP(x,y));
end
end
end
% H1 = [-1 -1; 1 1];
% H2 = [1 -1; 1 -1];
% fx = conv2(I1,H1);
% fy = conv2(I1,H2);
% f = sqrt(fx.^2 + fy.^2);
% a = atan(fx./eps(fy));
for i = 1:m
for j = 1:n
      if(a(i,j) > -pi/8 && a(i,j) <= pi/8)
         a(i,j) = 0;
      elseif(a(i,j) > pi/8 && a(i,j) <= 3*pi/8)
         a(i,j) = 1;
      elseif((a(i,j) > 3*pi/8 || a(i,j) <= -3*pi/8))
         a(i,j) = 2;
      else
         a(i,j) = 3;
      end
end
end
%（3）对梯度幅值进行非极大值抑制；
F = f;
for i =2

m-1)
for j=2

n-1)
      switch(a(i,j))
         case 0
            if(F(i,j) <= F(i,j-1) || F(i,j) <= F(i,j+1))
                  f(i,j) = 0;
            end
         case 1
            if(F(i,j) <= F(i-1,j+1) || F(i,j) <= F(i+1,j-1))
                  f(i,j) = 0;
            end
         case 2
               if(F(i,j) <= F(i-1,j) || F(i,j) <= F(i+1,j))
                  f(i,j) = 0;
               end
         case 3
               if(F(i,j) <= F(i-1,j-1) || F(i,j) <= F(i+1,j+1))
                  f(i,j) = 0;
               end
      end
end
end
figure,imshow(f),title('非极大值抑制后结果');
%step4:用双阈值算法检测和连接边缘。
t1 =str2double(get(handles.edit1,'string'));%低阈值
t2 =str2double(get(handles.edit2,'string'));%高阈值
Canny_1 = zeros(m,n);
Canny_2 = zeros(m,n);
for i = 1:m
for j =1:n
      if(f(i,j) >= t1 )
         Canny_1(i,j) = 255;
      else
         Canny_1(i,j) = 0;
      end
      if(f(i,j) >= t2 )
         Canny_2(i,j) = 255;
      else
         Canny_2(i,j) = 0;
      end
end
end

for i = 2

m-1)
for j = 2

n-1)
      if((Canny_1(i,j)==255) && (Canny_2(i,j)~=255))
         if((Canny_2(i-1,j-1)==255) || (Canny_2(i-1,j)==255) || (Canny_2(i-1,j+1)==255)|| ( Canny_2(i,j-1)==255 ) || (Canny_2(i,j+1)==255) || (Canny_2(i+1,j-1)==255) || (Canny_2(i+1,j)==255) ||(Canny_2(i+1,j+1) ==255))
            Canny_2(i,j)=255;
         end
      end
end
end
Canny=Canny_2;
title('Canny算子提取的边缘');

回复此楼

» 猜你喜欢

1楼 2014-12-28 17:19:07

已阅回复此楼关注TA 给TA发消息送TA红花 TA的回帖

智能机器人

Robot (super robot)

我们都爱小木虫

找到一些相关的精华帖子，希望有用哦~

【求助】vc6.0运行一个exe文件对图像进行canny边缘检测时就会出现以下错误，请教如何� 已经有5人回复

点击这里搜索更多相关资源

科研从小木虫开始，人人为我，我为人人

相关版块跳转我要订阅楼主汤和加菲的主题更新

返回列表