用MATLAB写基于腐蚀和膨胀的预处理用分水岭法做边缘检测

以下是MATLAB代码实现基于腐蚀和膨胀的预处理用分水岭法做边缘检测的步骤：

1. 读取图像并将其转换为灰度图像

img = imread('image.jpg');
gray_img = rgb2gray(img);

2. 对灰度图像进行高斯滤波

filtered_img = imgaussfilt(gray_img);

3. 对滤波后的图像进行腐蚀和膨胀操作

se = strel('disk', 3);
eroded_img = imerode(filtered_img, se);
dilated_img = imdilate(filtered_img, se);

4. 计算梯度幅值

gradient_img = imsubtract(dilated_img, eroded_img);

5. 对梯度幅值进行阈值处理

threshold = graythresh(gradient_img);
binary_img = imbinarize(gradient_img, threshold);

6. 对二值化图像进行连通区域分析并进行分水岭变换

labeled_img = bwlabel(binary_img);
dist_transform = bwdist(~binary_img);
watershed_img = watershed(dist_transform);

7. 将分水岭变换结果与原始图像进行叠加，得到边缘检测结果

edge_img = gray2rgb(gray_img);
edge_img(watershed_img == 0) = [255, 0, 0];

完整代码如下：

img = imread('image.jpg');
gray_img = rgb2gray(img);
filtered_img = imgaussfilt(gray_img);
se = strel('disk', 3);
eroded_img = imerode(filtered_img, se);
dilated_img = imdilate(filtered_img, se);
gradient_img = imsubtract(dilated_img, eroded_img);
threshold = graythresh(gradient_img);
binary_img = imbinarize(gradient_img, threshold);
labeled_img = bwlabel(binary_img);
dist_transform = bwdist(~binary_img);
watershed_img = watershed(dist_transform);
edge_img = gray2rgb(gray_img);
edge_img(watershed_img == 0) = [255, 0, 0];
imshow(edge_img);