How to segment cursive/connected (Arabic) word into characters?

I want to segment the connected character. Based on the histogram/profile, i assume that I could segment the character based on it's baseline. but, I still couldn't built an appropriate code to make it works.

I hope, anyone could help :)

The following is the code that could segment isolated character very well. But, still not work for connected character.

% // Original Code by Soumyadeep Sinha //
  % Saving each  single segmented character as one file 

    function [segm] = trysegment (a)
    myFolder = 'D:\1. Thesis FINISH!!!\Data set\trial';
    level = graythresh (a);
    bw = im2bw (a, level);
    b = imcomplement (bw);
    i= padarray(b,[0 10]);
    verticalProjection = sum(i, 1);
    set(gcf, 'Name', 'Trying Segmentation for Cursive', 'NumberTitle', 'Off') 
    subplot(2, 2, 1);imshow(i); 
    subplot(2,2,3);
    plot(verticalProjection, 'b-'); %histogram show by this code
    % hist(reshape(input,[],3),1:max(input(:))); 
    grid on;
    % % t = verticalProjection;
    % % t(t==0) = inf;
    % % mayukh = min(t)
    % 0 where there is background, 1 where there are letters
    letterLocations = verticalProjection > 0; 
    % Find Rising and falling edges
    d = diff(letterLocations);
    startingColumns = find(d>0);
    endingColumns = find(d<0);
    % Extract each region
    y=1;
    for k = 1 : length(startingColumns)
      % Get sub image of just one character...
      subImage = i(:, startingColumns(k):endingColumns(k)); 
    %   se = strel('rectangle',[2 4]);
    %   dil = imdilate(subImage, se); 
      th = bwmorph(subImage,'thin',Inf);
      n = imresize (th, [64 NaN], 'bilinear');
      figure, imshow (n);  
    [L,num] = bwlabeln(n);
    for z= 1 : num
    bw= ismember(L, z);
    % Construct filename for this particular image.
    baseFileName = sprintf('char %d.png', y);
    y=y+1;
    % Prepend the folder to make the full file name.
    fullFileName = fullfile(myFolder, baseFileName);
    % Do the write to disk.
    imwrite(bw, fullFileName);
    % subplot(2,2,4);
    % pause(2);
    % imshow(bw);
    end
    % y=y+1;
    end;
    segm = (n);

This is the sample of image.

%

[Sivakumaran Chandrasekaran]

try ocr as well as number plate extraction method. it may help

[ana ain]

Hi, Sivakumaran

Could you provide an appropriate code? Or please explore more about your suggestion.

Thank you so much :)

[Walter Roberson]

"number plate extraction" is something you could do on arabic letters if you were able to do the arabic ocr that is being looked for here. You would need to be able to do this task first. Likewise, ocr is what you are trying to do here.

[ana ain]

Walter and Sivakumaran

Yes, i am trying to do ocr, but still stuck in this segmentation stage.

Thank you so much for the suggestion. I'll try it.

[Ainatul Radhiah]

Hi Ana , Can I get coding for the segmentation of the Arabic alphabet that has been successful? I need for my thesis, thank you very much, this is my email [e-mail]

[Nina Angraheni]

Hi Ana , Can I get coding for the segmentation of the Arabic alphabet that has been successful? I need for my skripsi, thank you very much, this is my email [e-mail]

[hatem herzawy]

can i get the project on my Email please [e-mail]

[RaeD AL-JaaFari]

Hi Ana, could I get the project on my Email address, please [e-mail]

[Mary Hamad]

can I have the code in my email address because I'm trying to develop OCR for Arabic and I also stuck here :( [e-mail]

[Mohamed sidhoum]

Hey ana,

I want just a help. I saw in ResearchGate that a professor sent you the method on how to segment an Arabic handwriitng word with MATLAB, (with each letter of the word in a picture). Can please email me? I need the code for my mastery final project and I'm stuck.

email : [e-mail]

Have a nice day .

[Image Analyst]

Mohamed, please see my attached demo for the Bengali language. Maybe it's similar enough for you to adapt.

Answer 1

[chandrapal Singh]

Blob analysis

Answer 2

[Image Analyst]

See http://www.visionbib.com/bibliography/contentschar.html#OCR,%20Document%20Analysis%20and%20Character%20Recognition%20Systems

23.4.8 Other Character Sets
  23.4.8.1 Devanagari, Indic, Hindi, Hindu, Bangla, Bengali, Telugu, Characters
  23.4.8.2 Arabic Character Recognition
  23.4.8.3 Arabic Recognition, Word Level, Word Spotting
  23.4.8.4 Farsi, Persian Character Recognition

Answer 3

[ih ih]

hi ana can get the code if u please

ineed ocr arbic to my help me in project

[e-mail]