How to find the distance between vectors whose coordinates are stored in different arrays.

Hi,

I get the error:??? Cell contents assignment to a non-cell array object. at this line in my code:

 for k=1:N 
 A{k}=zeros(length(m),2);
 end

That's my first problem. Assuming that I figure out how to solve my first problem, I have another problem: once I have stored values in A{N}, I would like to compute

A{1}(1,:)-A{2}(1,:), A{1}(2,:)-A{2}(2,:), A{1}(3,:)-A{2}(3,:), ..., A{1}(n,:)-A{2}(n,:)
A{2}(1,:)-A{3}(1,:), A{2}(2,:)-A{3}(2,:), A{2}(3,:)-A{3}(3,:), ... , A{2}(n,:)-A{3}(n,:)
A{3}(1,:)-A{4}(1,:), A{3}(2,:)-A{4}(2,:), A{3}(3,:)-A{4}(3,:), ... , A{3}(n,:)-A{4}(n,:)
.
.
.

A{N-1}(1,:)-A{N}(1,:), A{N-1}(2,:)-A{N}(2,:), A{N-1}(3,:)-A{N}(3,:),..., A{N-1}(n,:)-A{N}(n,:)

and store these values into an n X N matrix.

I know I'm gonna find some help here, I always do.

Please help me solve my two problems.

Thanks.

[Matt J]

Your example doesn't lead to an n X N matrix. Each A{i}(j,:) is a 1x2 vector, so this leads to a N-1 x 2*n matrix.

Answer 1

[Matt J]

Your first problem.

 A=cell(N,1);

 for k=1:N 
  A{k}=zeros(length(m),2);
 end

[Matt J]

Your 2nd problem

 f=@(c) reshape(c.',1,[]);

 A=cell2mat(cellfun(f,A(:),'uni',0));

 T=speye(N-1,N); 
   T(N:N:end)=-1;

 result = reshape( (T*A).', 2,[]);
 result = sqrt(sum(result.^2,1));
 result = reshape(result,n,[]);

[MatlabFan]

Thank you very much Matt. There are really few of you guys around. I do thank you for your time.

I get a bunch of zeros in result. I shouldn't have that. Below is the full code.

I am sure the error is where I used

      rgb=imread('test_image_%d.png', N);

to read the saved images.

How can I read the saved images?

 clc
[SigOrg, fs]=wavread('SpeakerO.wav');
fL = 30;        % filter low cutoff freqency
fH = 800;       % filter high cutoff frequency
FiltOrd = 4;    % filter order
[bl,al]=butter(FiltOrd,[fL fH]/(fs/2));   %creating the coefficients for a bandpass filter with low freq. fL and high freq. fH
sig=filter(bl,al,SigOrg);                 %filtering out frequencies below 30 and above 600 Hz from the signal

 sigCorr = xcorr(sig);
 [A,K] = findpeaks(sigCorr); % find the peaks and their locations in the  cross correlation
 [max1 indx1] = max(A);  % find the highest peak and its location among the peaks of the cross correlation
 loc1 = K(indx1);    % loc1 is the location of the highest peak in the % cross correlation
[A2, K2] = findpeaks(A); % find the peaks and their location among the peaks of the cross correlation's peaks
[max2 indx2] = max(A2);
loc2 = K2(indx2+1);
loc3 = K(loc2);
Pitch = fs/(loc3-loc1);
Num=floor(length(SigOrg)/(loc3-loc1));
x=1;
for N=1:Num
imtool close all;
set(gcf,'Visible','off');
 plot(SigOrg(x:N*(loc3-loc1)))
 x=N*(loc3-loc1);
 filename = sprintf('test_image_%d.png', N);
saveas(gcf, filename, 'png') 
end
 A=cell(1,Num);
for N=1:Num

 rgb=imread('test_image_%d.png', N);
 imshow(rgb)

 r=rgb(:,:,1);
g=rgb(:,:,2);
b=rgb(:,:,3);
[i j] = find(r==0 & g==0 & b==255);

 Counts = accumarray(j,1);
 Counts=Counts(Counts~=0);
 length(Counts)
Vec=accumarray(j,i);
Vec=Vec(Vec~=0);
m=floor(Vec./Counts);
n=j(1):j(end)
A{N}=zeros(length(m),2);
length(m);
length(n);

 A{N}(:,1)=n;
 A{N}(:,2)=m;
 end
 clc
 f=@(c) reshape(c.',1,[]);
 A=cell2mat(cellfun(f,A(:),'uni',0));
 T=speye(Num-1,Num); 
   T(Num:Num:end)=-1;
 result = reshape( (T*A).', 2,[]); % --------Please what does this do?
 result = sqrt(sum(result.^2,1)); % contains the distance between vectors in the same position but different figure.
 result = reshape(result,length(n),[])% ------Please what does this do?
  end

[MatlabFan]

Ok, I think that I figured out how to read the saved images. I applied the following:

 rgb=cell(1,Num);
 for N=1:Num
 imtool close all;
 set(gcf,'Visible','off');
  plot(SigOrg(x:N*(loc3-loc1)))
  x=N*(loc3-loc1);
  filename = sprintf('test_image_%d.png', N);
 saveas(gcf, filename, 'png')
  rgb{N}=imread(filename);

 imshow(rgb{N})
end

My problem is that it does not display imshow. It says the following when I type, say, imshow(rgb{4}):

Warning: Image is too big to fit on screen; displaying at 50% > In imuitools\private\initSize at 73 In imshow at 262

Also, the code runs into the following error:

??? Error using ==> cat CAT arguments dimensions are not consistent.

Error in ==> cell2mat at 85 m{n} = cat(1,c{:,n});

Error in ==> ImagExperiment at 55 A=cell2mat(cellfun(f,A(:),'uni',0));

Thank you Matt, You help me a lot.

Would anybody else assist me, please?

Thanks.

[Matt J]

Errors like these are a job for the debugger. Set a breakpoint at line 55 in ImagExperiment and rerun your code. When the code halts at line 55, run

 z=cellfun(@(c) size(c), A,'uni',0); z{:}

at the K>> prompt and show us the output.

[MatlabFan]

Hi Matt, I found a way around this. And yes, I thank you for your great help. Now, I am experiencing other issues with my code.