How can I determent baseline in arabic handwriting

I have problem in detect Arabic handwriting baseline . first i I used nested forloop to sum the black pixel in each line ,then the row which have max value it is baseline . so some tel me use histogram,but i don't now how to determend .

Image  = imread('40-1.png');

%-----------------------------------------------
imshow(d);
BW  = im2bw(d,0.8);

nelements = hist(BW)

   [row,col]=size(BW)
   %_______________________________
  z=plot(BW)
  %***********************************
  c=0;
  counts=0;
  cou=0;
  for i=1:row
      for j=1:col

          if (bw(i,j)==0)
              c(i)=c(i)+1
          end
      end
  s(i)=c(i)

end
V=max(s)

Answer 1

[Image Analyst]

For successful algorithms, see http://iris.usc.edu/Vision-Notes/bibliography/contentschar.html#OCR,%20Document%20Analysis%20and%20Character%20Recognition%20Systems

23.4.8.2 Arabic Character Recognition
23.4.8.3 Arabic Recognition, Word Level, Word Spotting

[ra]

(The row containing the largest run of black pixels in the word image is located) How ??. So, how to use vertical histogram of pixels in the image is analysed please help me.. code

[Image Analyst]

verticalProfile = sum(grayImage, 2);

Then threshold

darkRows = verticalProfile < 3000; % or whatever.

That will tell you what lines have black writing on them.

[ra]

My image is Binary Image when run this is appears .How to detect the max row 166 166 166 166 166 166 166 166 166 166 166 166 166 166 166 166 166 166 166 166 166 166 166 166 164 164 162 162 162 162 162 162 162 162 162 162 146 146 140 140 146 146 140 140 132 132 124 124 122 122 124 124 126 126 136 136 140 140 140 140 138 138 142 142 150 150 150 150 158 158 156 156 166 166 166 166 166 166 166 166 166 166 166 166 166 166 166 166

[Image Analyst]

What does "max row" mean?

[ra]

The row have maximum number of black pixel is baseline of handwriting

[Image Analyst]

[darkestSum, maxRow] = min(yourProfileArray);