how to find the location of element in matrix?
    7 Ansichten (letzte 30 Tage)
  
       Ältere Kommentare anzeigen
    
    sheno39
 am 18 Okt. 2013
  
    
    
    
    
    Kommentiert: Azzi Abdelmalek
      
      
 am 28 Okt. 2013
            my code:
I=[2 3 10 4 6; 1 4 7 5 3; 5 2 8 4 3;8 2 1 7 3;1 9 8 3 4;];
D = padarray(I,[1 1],0,'both');
[x y]=size(D);
m=1;
n=1;
for i=2:x-1
    for j=2:y-1
        A=[D(i-1,j-1:j+1),D(i,j-1),D(i,j+1),D(i+1,j-1:j+1)];
        I1(m,n)=max(A(:));
        %[maxval(m,n) index(m,n)]=max(A(:));
        n=n+1;
    end
    m=m+1;
    n=1;
end
[tc locatn]=ismember(I1,I);
s=[5,5];
[R C]=ind2sub(s,locatn');
Required output
locatn=[7 11 12 11 17; 3 11 11 11 21; 4 4 12 13 19; 10 10 10 13 19; 10 4 10 15 19]
i got the output as,
locatn=[7 11 12 11 3; 3 11 11 11 21; 4 4 12 4 12; 10 10 10 4 12; 10 4 10 4 12]
can anyone help me to get the required output? how to correct this code?
0 Kommentare
Akzeptierte Antwort
  Andrei Bobrov
      
      
 am 18 Okt. 2013
        
      Bearbeitet: Andrei Bobrov
      
      
 am 18 Okt. 2013
  
      I=[2 3 10 4 6; 1 4 7 5 3; 5 2 8 4 3;8 2 1 7 3;1 9 8 3 4;];
D = padarray(I,[1 1],0,'both');
s = size(I);
N = padarray(reshape(1:numel(I),s),[1 1],0,'both');
pt = [1 1 1;1 0 1;1 1 1] > 0;
s1 = s+2;
out = zeros(s);
for ii=2:s1(1)-1
    for jj=2:s1(2)-1
        D1 = D(ii-1:ii+1,jj-1:jj+1);
  K = N(ii-1:ii+1,jj-1:jj+1);
  K = K(pt);
        [~,i0] = max(D1(pt));
  out(ii-1,jj-1) = K(i0);
    end
end
or
I=[2 3 10 4 6; 1 4 7 5 3; 5 2 8 4 3;8 2 1 7 3;1 9 8 3 4;];
D = padarray(I,[1 1],0,'both');
s = size(I);
pt = [1 1 1;1 0 1;1 1 1] > 0;
a = -1:1;
j1 = [1;1;1]*a;
i1 = j1';
i1 = i1(pt);
j1 = j1(pt);
s1 = s+2;
R = zeros(s);
C = zeros(s);
for ii=2:s1(1)-1
    for jj=2:s1(2)-1
        D1 = D(ii-1:ii+1,jj-1:jj+1);
        [~,i0] = max(D1(pt));
        R(ii-1,jj-1) = ii + i1(i0) - 1;
        C(ii-1,jj-1) = jj + j1(i0) - 1;
    end
end
or
I=[2 3 10 4 6; 1 4 7 5 3; 5 2 8 4 3;8 2 1 7 3;1 9 8 3 4;];
Ip = padarray(I,[1 1],0,'both');
s = size(Ip);
s0 = s - 2;
idxi = padarray(reshape(1:numel(I),s0),[1 1],0,'both');
ptr = reshape(bsxfun(@plus,(0:2)',(0:2)*s(1)),1,[]);
p2 = bsxfun(@plus,(1:s0(1))',(0:s0(2)-1)*s(1));
idx = bsxfun(@plus,p2(:),ptr([1:4,6:end]));
[~,ii] = max(Ip(idx),[],2);
s1 = size(idx);
out = reshape(idxi(idx(sub2ind(s1,(1:s1(1))',ii))),s0);
[R,C] = ind2sub(size(out),out);
0 Kommentare
Weitere Antworten (1)
  Azzi Abdelmalek
      
      
 am 18 Okt. 2013
        
      Bearbeitet: Azzi Abdelmalek
      
      
 am 18 Okt. 2013
  
      I think there is an error in your expected locatn, try this
I2=unique(I1(:));
ii=histc(I1(:),I2);
idx1=zeros(size(I));
for k=1:numel(I2)
  a=I2(k)
  jdx=ismember(I1,a);
  j1=find(jdx,ii(k))';
  idx=ismember(I,a);
  i1=find(idx,ii(k));
  i1(end:numel(j1))=i1(end);
  idx1(j1)=i1;
end
locatn=idx1
2 Kommentare
  Azzi Abdelmalek
      
      
 am 28 Okt. 2013
				If you look for the location of 5 from I1 in the matrix I, there are two 5 in I1, we should find the locations of the two first 5 in I, that's what i1 represent.
Siehe auch
Kategorien
				Mehr zu Logical finden Sie in Help Center und File Exchange
			
	Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!


