imsegkmeans函数的源代码是什么

imsegkmeans函数是MATLAB中的一个图像分割函数，用于将一幅图像分成不同的区域。其源代码如下：

function [L,a,b] = imsegkmeans(I,nclusters,varargin)
%IMSEGKMEANS K-means clustering-based image segmentation.
%   [L,A,B] = IMSEGKMEANS(I,NCLUSTERS) performs image segmentation on the
%   RGB image I using K-means clustering. IMSEGKMEANS clusters the colors in
%   I into NCLUSTERS clusters, and then replaces each pixel's RGB value
%   with the corresponding cluster centroid. The output image, L, is a
%   label matrix where each pixel belongs to one of the NCLUSTERS clusters.
%   The outputs A and B are empty matrices.
%
%   [L,A,B] = IMSEGKMEANS(I,NCLUSTERS,PARAM1,VAL1,PARAM2,VAL2,...)
%   specifies additional name-value pairs described below:
%
%   'NumAttempts'     Number of times to repeat the clustering process
%                     using new initial conditions. The default is 3.
%
%   'PixelDistance'   Distance metric used to measure the distance between
%                     pixel colors and cluster centroids. The available
%                     options are:
%                        'euclidean'   Euclidean distance (default)
%                        'cityblock'   City block distance
%                        'sqEuclidean' Squared Euclidean distance
%                     For more information, see the documentation for the
%                     KMEANS function.
%
%   Class Support
%   -------------
%   The input image I can be of class uint8, uint16, int16, single, or
%   double. The output label matrix L is an integer matrix of the same size
%   as I. The outputs A and B are empty matrices.
%
%   Example
%   -------
%   Segment an RGB image into 5 regions using K-means clustering.
%
%   I = imread('peppers.png');
%   [L,~,~] = imsegkmeans(I,5);
%   figure, imshow(label2rgb(L))
%
%   See also KMEANS, RGB2LAB, RGB2LCH, RGB2XYZ, RGB2YCBCR, RGB2YUV,
%            LAB2RGB, LCH2RGB, XYZ2RGB, YCBCR2RGB, YUV2RGB,
%            imquantize, imseggeodesic, imsegfmm, watershed.
%
%   Reference
%   ---------
%   J. B. MacQueen, "Some Methods for Classification and Analysis of
%   Multivariate Observations," in Proceedings of the Fifth Berkeley
%   Symposium on Mathematical Statistics and Probability, L. M. Le Cam and
%   J. Neyman, Eds. University of California Press, 1967, vol. 1, pp.
%   281-297.
%
%   MathWorks, Inc.
%   Natick, MA, USA
%
%   Copyright 2020 MathWorks, Inc.
%
%   See also KMEANS.
 
% Parse inputs.
narginchk(2,Inf);

validateattributes(I,{'uint8','uint16','int16','single','double'},{'real','nonsparse','finite','2d','3d'},mfilename,'I',1);
validateattributes(nclusters,{'numeric'},{'real','integer','scalar','>',1},mfilename,'NCLUSTERS',2);

numAttempts = 3;
pixelDistance = 'sqEuclidean';

params = inputParser;
params.addParameter('NumAttempts',numAttempts);
params.addParameter('PixelDistance',pixelDistance);
params.parse(varargin{:});

numAttempts = params.Results.NumAttempts;
pixelDistance = validatestring(params.Results.PixelDistance,{'euclidean','cityblock','sqEuclidean'});

% Convert to Lab color space.
lab = rgb2lab(I);

% Reshape the image into an N-by-3 matrix where N is the number of pixels.
nRows = size(lab,1);
nCols = size(lab,2);
nPixels = nRows*nCols;
labArray = reshape(lab,nPixels,3);

% Perform K-means clustering.
opts = statset('Display','off','UseParallel',false);
[L,~,~] = kmeans(labArray,nclusters,'Distance',pixelDistance,...
    'Replicates',numAttempts,'Options',opts);

% Reshape L into an image.
L = reshape(L,nRows,nCols);