Life Beyond Code

My Research Publication in IEEE, "Language Independent OCR for handwritten text"

Posted on: January 11, 2007

For my graduation research work, i got my first reseach pubication in IEEE in December 2004. The title of my paper is “Language Independent Optical Character Recognition(OCR) for handwritten text”. If you have a valid IEEE login , you can access this research paper at following link.

Project Abstract:


Handwritten Language Independent Optical Character Reader (OCR) has emerged as a new idea to overcome the difficulties to understand the document datasets for different languages in single OCR software. Main idea behind Handwritten Language independent OCR is that all languages can be represented in some common basic geometrical strokes. Our research show that, with stroke based character segmentation and a Neural Network trained on basic geometrical shapes, a Language independent OCR can be born.

This software was developed in C#.NET and it is for windows operating system.

This software uses Pitman Shorthand Language and English as its testing languages. Shorthand is an international language used by journalists/stenographers in live conferences and seminars because of its high speed recording advantage (can record over 200-1200 words/min.).As shorthand is written with lead pencil, it has a poor document image with very less gray level difference between background and original text.

This software takes as input a handwritten document, perform necessary operations to convert document into a form suitable for further processing. After preprocessing subsequent operations involve, removing background lines ( Hough Transformation ) and noise, finding connected components , produce skeleton of each component, segmenting individual components into its basic geometrical strokes , extracting necessary features from these strokes, classification and recognition of PSL characters by analyzing these strokes using a trained Neural Network for basic geometrical shapes, and finally converts the recognized PSL strokes to editable form.

Being Language independent OCR, it covers wide areas of applications. This software can be perfectly used for encryption because user can define his own language too. It can also be used for data updates of offline literature written in any language, for enhancing the idea of paper-less office, for making a newspaper reader for blinds. It can be used by newspaper agencies for translating their journalist’s hand written PSL documents into Standard English requiring no more effort of stenographers for understanding PSL documents.

1 Response to "My Research Publication in IEEE, "Language Independent OCR for handwritten text""

Bush is forever saying that democracies do not invade other countries and start wars. Well, he did just that. He invaded Iraq, started a war, and killed people. What do you think? Why has bush turned our country from a country of hope and prosperity to a country of belligerence and fear.
Are we safer today than we were before?
The more people that the government puts in jails, the safer we are told to think we are. The real terrorists are wherever they are, but they aren’t living in a country with bars on the windows. We are.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


January 2007
« Sep   Feb »


RSS Readings

  • An error has occurred; the feed is probably down. Try again later.

And Here is ME...

%d bloggers like this: