[Project-ideas] Project-Printed Bangla OCR System.

[Sankarshan Mukhopadhyay]

On Mon, Apr 29, 2013 at 4:09 PM, gaurav bhadra <gaurav.bhadra at gmail.com> wrote:
> I am a final year student of St. Xavier's College, Kolkata pursuing B.Sc in
> Computer Science. I will be finishing my bachelor's course in a few months
> and will be pursuing my Master's degree from the University of Oxford. Since
> my Master's course starts in October, I have a lot of free time now and
> would like to take this opportunity to work on a project for Google Summer
> Of Code.

I hope you have thought long and hard about this. The relocation to a
different city for studies is a bit of a time sink and, even though
you foresee a lot of time on your hands, niggling logistics and other
administrative aspects tend to soak them up.

The above statement is not meant to discourage you - it is something I
have seen again and again.

> I would like to work on Printed Bangla OCR System. It is an interesting
> topic and will be a nice challenge improving the performance of current
> systems to above 98%! I will be submitting a detailed proposal by tomorrow
> specifying how I have planned to implement it. I like the initiative taken
> by Ankur.org.in and the ideology of the organization. Since my mother tongue
> is Bangla, it gives me an added impetus to work on this project. Though my
> project will mainly focus on implementing a Bangla OCR system, it should
> provide a basic framework for other Indic languages.

While you work on your proposal, it would be nice if you can also
catch up with the discussions on this list (using the archives) and
check if all your doubts are clarified.

> I have worked on related topics such as Image Processing and Pattern
> Matching in MATLAB and Java. I am also proficient in SQL, C and C++.

Have you pushed any of your code into Github or, like repositories?



--
sankarshan mukhopadhyay
<https://twitter.com/#!/sankarshan>