reduce the extended characters in image

how are you everyone

I need your help to solve this problem :

how can I reduce the elongated parts in this word , this elongated parts do not have any meaning so I need to delete them .

As we have below in the image, the property of these parts is that every part is located at the same number of rows ,there are no pixels above or below .

any answer will be appreciated !

thank you

Answer 1

[Image Analyst]

I suggest you get an algorithm from the experts who have already figured it out and published it, because there's nothing in MATLAB to do that at a high level. So find a paper here http://www.visionbib.com/bibliography/contentschar.html#OCR,%20Document%20Analysis%20and%20Character%20Recognition%20Systems

Look especially at these sections:

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

Then pick a paper, and code it up, or ask the authors if you can buy their code.

[majed majed]

Thank you for your answering !

I have looked the links but I didn't find any answer . We can consider the problem as next :

If We have a vector with n elements , Those elements are repeated at some places , how we can rewrite this vector with just two times for the repeated elements and leave others as they are . For example if we have V=[4444444432566666611111111222222222234]

The wished result is

VV=[443256611223] Thank you .

[majed majed]

I got it

Thank you !

[Image Analyst]

Wow, you're a fast reader. You're welcome. Looks like you can get a paper out of it since not one of all the other published papers on it were able to solve it.