Matrix dimension must agree error in the code,need a solution

function [clusters, result_image, clusterized_image] = kmeanssegment(im,k)

%histogram calculation
hist_value = zeros(256,1);
img_hist = zeros(1,256); % Reset
tic
[rows, columns] = size(im);
for col = 1 : columns
  for row = 1 : rows
    img_hist(im(row,col)+1) = img_hist(im(row,col)+1) + 1;
  end
end
toc
for i=1:256
  hist_value(i)=i-1; %#ok<NASGU>
end
%cluster initialization
cluster=zeros(k,1);
cluster_count = zeros(k,1);
for i=1:k
  cluster(i)=uint8(rand*255);
end;

old = zeros(k,1);
while (sum(sum(abs(old-cluster))) >k)
  old = cluster;
  closest_cluster = zeros(256,1);
  min_distance = uint8(zeros(256,1)); %#ok<NASGU>
  min_distance = abs(hist_value-cluster(1));

  %calculate the minimum distance to a cluster
  for i=2:k
    min_distance =min(min_distance,  abs(hist_value-cluster(i)));
  end;

  %calculate the closest cluster
for i=1:k
    closest_cluster(min_distance==(abs(hist_value-cluster(i)))) = i;
end
  %calculate the cluster count
  for i=1:k
    cluster_count(i) = sum(img_hist.*(closest_cluster==i));
  end;
for i=1:k
if (cluster_count(i) == 0)
      cluster(i) = uint8(rand*255);
    else
      cluster(i) = uint8(sum(img_hist(closest_cluster==i).*hist_value(closest_cluster==i))/cluster_count(i));
end
end
  end
imresult=uint8(zeros(size(im)));
for i=1:256
  imresult(im==(i-1))=cluster(closest_cluster(i));
end;

clustersresult=uint8(zeros(size(im)));
for i=1:256
  clustersresult(im==(i-1))=closest_cluster(i);
end;

clusters = cluster;
result_image = imresult;
clusterized_image = clustersresult;
end

Error Elapsed time is 0.008945 seconds. ??? Error using ==> times Matrix dimensions must agree.

Error in ==> kmeansclustering at 52 cluster(i) = uint8(sum(img_hist(closest_cluster==i).*hist_value(closest_cluster==i))/cluster_count(i));

Error in ==> kmeansclusteringdemo at 82 k=kmeansclustering(im,5);

Answer 1

[Image Analyst]

Standard debugging technique is to break up the long expressions into smaller parts and look at them, so

sum(img_hist(closest_cluster==i).*hist_value(closest_cluster==i))

becomes

a=img_hist(closest_cluster==i)
whos a
b = hist_value(closest_cluster==i)
whos b
c = sum(a.*b)
whos c

Give that a try and see what you learn. a and b should be the same size to do the element by element multiplication. But they won't be - that's why you got the error.