改进canny算法matlab代码效果图

很抱歉，作为AI语言模型，我无法提供图片。但是，我可以为您提供改进Canny算法的Matlab代码。

改进Canny算法的Matlab代码：

function edgeImg = myCanny(im, sigma, thLow, thHigh)
%MYCANNY Implementation of Canny edge detection algorithm
%   edgeImg = MYCANNY(im, sigma, thLow, thHigh) returns binary edge map
%   of input image 'im' using Canny edge detection algorithm. 'sigma' is
%   the standard deviation of Gaussian filter used for smoothing. 'thLow'
%   and 'thHigh' are thresholds for hysteresis thresholding.

% Convert image to double
im = im2double(im);

% Smoothing using Gaussian filter
G = fspecial('gaussian', [5 5], sigma);
imFilt = imfilter(im, G, 'same');

% Computing gradients using Sobel filter
Sx = [-1 0 1; -2 0 2; -1 0 1];
Sy = [-1 -2 -1; 0 0 0; 1 2 1];
Gx = imfilter(imFilt, Sx, 'same');
Gy = imfilter(imFilt, Sy, 'same');
Gmag = sqrt(Gx.^2 + Gy.^2);
Gtheta = atan2(Gy, Gx) * 180/pi;

% Non-maximum suppression
[n, m] = size(im);
edgeImg = zeros(n, m);
for i = 2:n-1
    for j = 2:m-1
        if (Gtheta(i,j) < 0)
            Gtheta(i,j) = Gtheta(i,j) + 180;
        end
        
        if ((Gtheta(i,j) >= 0) && (Gtheta(i,j) < 22.5) || ...
                (Gtheta(i,j) >= 157.5) && (Gtheta(i,j) < 180))
            q = Gmag(i, j+1);
            r = Gmag(i, j-1);
        elseif ((Gtheta(i,j) >= 22.5) && (Gtheta(i,j) < 67.5))
            q = Gmag(i+1, j-1);
            r = Gmag(i-1, j+1);
        elseif ((Gtheta(i,j) >= 67.5) && (Gtheta(i,j) < 112.5))
            q = Gmag(i+1, j);
            r = Gmag(i-1, j);
        elseif ((Gtheta(i,j) >= 112.5) && (Gtheta(i,j) < 157.5))
            q = Gmag(i-1, j-1);
            r = Gmag(i+1, j+1);
        end
        
        if (Gmag(i,j) >= q) && (Gmag(i,j) >= r)
            edgeImg(i,j) = Gmag(i,j);
        end
    end
end

% Hysteresis thresholding
thLow = thLow * max(max(edgeImg));
thHigh = thHigh * max(max(edgeImg));
for i = 1:n
    for j = 1:m
        if (edgeImg(i,j) < thLow)
            edgeImg(i,j) = 0;
        elseif (edgeImg(i,j) > thHigh)
            edgeImg(i,j) = 1;
        else
            if ((edgeImg(i-1,j-1) > thHigh) || (edgeImg(i-1,j) > thHigh) || ...
                    (edgeImg(i-1,j+1) > thHigh) || (edgeImg(i,j-1) > thHigh) || ...
                    (edgeImg(i,j+1) > thHigh) || (edgeImg(i+1,j-1) > thHigh) || ...
                    (edgeImg(i+1,j) > thHigh) || (edgeImg(i+1,j+1) > thHigh))
                edgeImg(i,j) = 1;
            end
        end
    end
end

% Convert edge map to binary image
edgeImg = im2bw(edgeImg);
end

使用方法：

im = imread('lena.jpg');
sigma = 1.5;
thLow = 0.05;
thHigh = 0.15;
edgeImg = myCanny(im, sigma, thLow, thHigh);
imshow(edgeImg);

其中，'lena.jpg'是您需要读取的图像文件名，sigma是高斯滤波器的标准差，thLow和thHigh是用于滞后阈值的阈值。您可以根据需要调整这些参数以获得最佳结果。