写一个matlab代码，利用Steger算法可以从激光条纹图像中提取激光条纹中心线，并且得到激光条纹中心线特征点坐标

以下是一个基于Steger算法的MATLAB代码，用于从激光条纹图像中提取激光条纹中心线，并获取激光条纹中心线特征点坐标。

function [centerline, feature_points] = StegerAlgorithm(I, sigma, threshold, min_distance)
% I：激光条纹图像
% sigma：高斯滤波器的标准差
% threshold：非极大值抑制的阈值
% min_distance：特征点之间的最小距离

% Apply Gaussian smoothing
G = fspecial('gaussian', round(3*sigma)*2+1, sigma);
I_smooth = imfilter(I, G, 'replicate');

% Compute partial derivatives
Ix = imfilter(I_smooth, [-1 0 1], 'replicate');
Iy = imfilter(I_smooth, [-1; 0; 1], 'replicate');
Ixx = Ix.^2;
Ixy = Ix.*Iy;
Iyy = Iy.^2;

% Apply Gaussian smoothing to derivatives
Gxx = imfilter(Ixx, G, 'replicate');
Gxy = imfilter(Ixy, G, 'replicate');
Gyy = imfilter(Iyy, G, 'replicate');

% Compute eigenvalues and eigenvectors of Hessian matrix
H = zeros(size(I,1), size(I,2), 2, 2);
D = zeros(size(I,1), size(I,2), 2);
for i = 1:size(I,1)
    for j = 1:size(I,2)
        H(i,j,:,:) = [Gxx(i,j) Gxy(i,j); Gxy(i,j) Gyy(i,j)];
        D(i,j,:) = eig(squeeze(H(i,j,:,:)));
    end
end

% Compute non-maximum suppression
N = size(I,1)*size(I,2);
is_max = zeros(N, 1);
for i = 1:N
    p = ceil(i/size(I,2));
    q = mod(i-1, size(I,2))+1;
    if D(p,q,1) > threshold && D(p,q,1) > D(p,q,2)
        if p > 1 && p < size(I,1) && q > 1 && q < size(I,2)
            H_sub = squeeze(H(p-1:p+1,q-1:q+1,:,:));
            H_sub = reshape(H_sub, [9,2,2]);
            [~,ind] = max(D(p,q,1)-squeeze(D_sub(:,1,:)));
            is_max(i) = ind == 5;
        else
            is_max(i) = true;
        end
    end
end

% Find feature points
feature_points = [];
[~,ind] = sort(D(:),'descend');
for i = 1:length(ind)
    if is_max(ind(i))
        [p,q] = ind2sub(size(I), ind(i));
        if ~any((feature_points(:,1)-q).^2 + (feature_points(:,2)-p).^2 < min_distance^2)
            feature_points = [feature_points; q p];
        end
    end
end

% Compute centerline
num_points = size(feature_points,1);
cost_matrix = zeros(num_points, num_points);
for i = 1:num_points
    for j = i+1:num_points
        cost_matrix(i,j) = sqrt(sum((feature_points(i,:)-feature_points(j,:)).^2));
        cost_matrix(j,i) = cost_matrix(i,j);
    end
end
[~, path] = dijkstra(cost_matrix, 1, num_points);
centerline = feature_points(path,:);

其中，`dijkstra`函数是用于计算最短路径的Dijkstra算法实现。