corr_alphabet1.m

人工神经网络资料

[alphabet,targets] =prprob1;
re=[];
%noisy = alphabet(:,i)+randn(35,1) * 0.4;
noisy=[0     1     1     1     1 ...
     0     1     0     0     1 ...
     0     1     0     0     1 ...
     0     1     0     0     1 ...
     0     1     0     0     1 ...
     0     1     0     0     1 ...
     0     1    1     1     1]';
for j=1:36;
Alp=alphabet(:,j);
%A=[Alp(1:5)';Alp(6:10)';Alp(11:15)';Alp(16:20)';Alp(21:25)';Alp(26:30)';Alp(31:35)'];
c=corrcoef(Alp,noisy);
re(i,j)=c(1,2);
end
[r,c]=max(re(i,:));
figure(1)
plotchar(noisy);
figure(2)
plotchar(alphabet(:,c));