function [XSelected,ySelected,XRest,yRest,vSelectedRowIndex]=ks_modification(X,y,Num)
% ks selects the samples XSelected which uniformly distributed in the exprimental data X's space
% Input
% X:the matrix of the sample spectra
% y:the matrix of the sample measurement values
% Num:the number of the sample spectra you want select
% Output
% XSelected:the sample spectras was selected from the X, i.e. the X of the calibration set
% ySelected:the sample measurement value was selected from the y, i.e. the y of the calibration set
% XRest:the sample spectras remain int the X after select, i.e. the X of the prediction set
% yRest:the sample measurement value remain int the y after select, i.e. the y of the prediction set
% vSelectedRowIndex:the row index of the selected sample in the X matrix
% Programmer: zhimin zhang @ central south university on oct 28 ,2007
% Modification: Penghui Liu @ Zhejiang University on Oct 10, 2022
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% start of the kennard-stone step one
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
nR=size(X,1); % obtain the size of the X matrix
mDistance=zeros(nR,nR); %dim a matrix for the distance storage
vAllofSample=1:nR;
for i=1:nR-1
vRX=X(i,:); % 获取X的一行数据
for j=i+1:nR
vRX1=X(j,:); % 获得X中的另一行数据
mDistance(i,j)=norm(vRX-vRX1); % 计算欧氏距离
end
end
[vMax,vIndexOfmDistance]=max(mDistance);
[~,nIndexofvMax]=max(vMax);
vSelectedSample(1)=nIndexofvMax;
vSelectedSample(2)=vIndexOfmDistance(nIndexofvMax);
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% end of the kennard-stone step one
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% start of the kennard-stone step two
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
for i=3:Num
vNotSelectedSample=setdiff(vAllofSample,vSelectedSample);
vMinDistance=zeros(1,nR-i+1);
for j=1:(nR-i+1)
nIndexofNotSelected=vNotSelectedSample(j);
vDistanceNew = zeros(1,i-1);
for k=1:(i-1)
nIndexofSelected=vSelectedSample(k);
if(nIndexofSelected<=nIndexofNotSelected)
vDistanceNew(k)=mDistance(nIndexofSelected,nIndexofNotSelected);
else
vDistanceNew(k)=mDistance(nIndexofNotSelected,nIndexofSelected);
end
end
vMinDistance(j)=min(vDistanceNew);
end
[~,nIndexofvMinDistance]=max(vMinDistance);
vSelectedSample(i)=vNotSelectedSample(nIndexofvMinDistance);
end
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%%%% end of the kennard-stone step two
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%%%% start of export the result
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
vSelectedRowIndex=vSelectedSample;
for i=1:length(vSelectedSample)
XSelected(i,:)=X(vSelectedRowIndex(i),:); %训练集光谱数据
ySelected(i,:)=y(vSelectedRowIndex(i),:); %训练集性质数据
end
vNotSelectedSample=setdiff(vAllofSample,vSelectedSample);
for i=1:length(vNotSelectedSample)
XRest(i,:)=X(vNotSelectedSample(i),:); %预测集光谱数据
yRest(i,:)=y(vNotSelectedSample(i),:); %预测集性质数据
end
end
