聚类matlab meanshift

以下是MeanShift的Matlab代码演示[^1]：

% Mean Shift Clustering Example
% Code written by Dr. Matthew E. Martin
% Assistant Professor, Department of Computer Science
% The University of Oklahoma
%
% This code demonstrates the Mean Shift clustering algorithm
%
% The data set used consists of 1000 points in 3D space.
% These points are divided into two clusters.
% One cluster is centered at position (30,30,30) and the other
% at position (80,80,80)
% Each point in the cluster is normally distributed with a standard
% deviation of 5 units.
%
% The Mean Shift clustering algorithm is then run on this data set and the
% resulting clusters are plotted using different colors for better visualization
%
% NOTE: This code is for educational purposes only and is not intended
% for commercial use without permission from the author.
%
% Code is provided "as is" and the author assumes no responsibility
% for any errors or problems that may arise from using this code.

% Create Data Set
x = [randn(1000,1)*5+30 randn(1000,1)*5+30 randn(1000,1)*5+30; ...
    randn(1000,1)*5+80 randn(1000,1)*5+80 randn(1000,1)*5+80];

% Implement Mean Shift
ms = MeanShift();
ms.bandwidth = 8;
ms.min_points = 10;
result = ms.cluster(x);

% Plot Results
figure;
hold on;
scatter3(result(:,1),result(:,2),result(:,3));
view(-115,40);

% Define MeanShift Class
classdef MeanShift
    properties
        bandwidth = 8;
        min_points = 10;
    end
    
    methods
        function cluster_result = cluster(obj,X)
            n = size(X,1);
            labels = zeros(n,1);
            cluster_center = [];
            visited = false(n,1);
            
            for i=1:n
                if ~visited(i)
                    visited(i) = true;
                    [new_cluster,labels] = obj.pointsInRange(X,X(i,:),visited);
                    while size(new_cluster,1) > 0
                        [new_cluster,labels2] = obj.pointsInRange(X,new_cluster(1,:),visited);
                        if size(new_cluster,1) >= obj.min_points
                            labels(labels2) = size(cluster_center,1)+1;
                            cluster_center = [cluster_center; mean(new_cluster)];
                        end
                        visited(labels2) = true;
                        new_cluster(1,:) = [];
                    end
                    if labels(i) == 0
                        cluster_center = [cluster_center; X(i,:)];
                        labels(i) = size(cluster_center,1);
                    end
                end
            end
            
            cluster_result = cluster_center(labels,:);
        end
        
        function [new_cluster,labels] = pointsInRange(obj,X,x,visited)
            distance = sqrt(sum((X-repmat(x,size(X,1),1)).^2,2));
            in_range = distance < obj.bandwidth;
            labels = find(in_range);
            new_cluster = X(in_range,:);
            new_cluster = new_cluster(~visited(in_range),:);
        end
    end
end

另外，还可以使用Python实现Mean Shift聚类算法，以下是Python代码示例[^2]：

import numpy as np
import matplotlib.pyplot as plt
from sklearn.cluster import MeanShift, estimate_bandwidth
from itertools import cycle

#Input dataset
X = np.array([[1, 2], [1, 4], [1, 0],
              [10, 2], [10, 4], [10, 0]])

#Estimate bandwidth
bandwidth = estimate_bandwidth(X, quantile=0.2, n_samples=500)

#Fit mean shift algorithm to data
ms = MeanShift(bandwidth=bandwidth, bin_seeding=True)
ms.fit(X)

#Extract cluster assignments for each data point
labels = ms.labels_

#Extract centroids
centroids = ms.cluster_centers_

#Number of clusters
n_clusters_ = len(np.unique(labels))

#Plot result
print("Number of estimated clusters : %d" % n_clusters_)
colors = cycle('bgrcmykbgrcmykbgrcmykbgrcmyk')
for k, col in zip(range(n_clusters_), colors):
    my_members = labels == k
    cluster_center = centroids[k]
    plt.plot(X[my_members, 0], X[my_members, 1], col + '.')
    plt.plot(cluster_center, cluster_center, 'o', markerfacecolor=col,
             markeredgecolor='k', markersize=14)
plt.title('Estimated number of clusters: %d' % n_clusters_)
plt.show()