# OCR project in Gsoc

**URL:** <https://www.ruby-forum.com/t/ocr-project-in-gsoc/163767>\
**Category:** Rails\
**Created:** [March 24, 2009, 6:50pm UTC](https://www.ruby-forum.com/t/ocr-project-in-gsoc/163767 "2009-03-24T18:50:41Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![Arulalan](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/a/8c91f0/32.png) [@Arulalan](https://www.ruby-forum.com/u/Arulalan)\
**Post date:** [March 24, 2009, 6:50pm UTC](https://www.ruby-forum.com/t/ocr-project-in-gsoc/163767/1 "2009-03-24T18:50:41Z")

</div>

Hi to all,

I planned to do project in gsoc… For OCR ( Optimal Character  
Recoganisation ) …

That is ,

```
         If we scanning one full text page from book, it will open

```

into open office as word format. so that we can edit the page from  
scanned text page… I planned to convert scanned letters to words for  
Tamil, English Languages… I will try to support few more languages  
also…This OCR project will can done by Using Rmagick , i will do  
this successfully.

```
         This is my idea, if any one of you can suggest me and

```

guide me to do this…

Thank,

Arulalan.

---

<div class="post-metadata">

**Author:** ![Arulalan](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/a/8c91f0/32.png) [@Arulalan](https://www.ruby-forum.com/u/Arulalan)\
**Post date:** [March 24, 2009, 10:53pm UTC](https://www.ruby-forum.com/t/ocr-project-in-gsoc/163767/2 "2009-03-24T22:53:36Z")

</div>

Hi,

What about Google Tesseract???

> **[tesseract-ocr](https://github.com/tesseract-ocr/)**
>
> Tesseract OCR. tesseract-ocr has 14 repositories available. Follow their code on GitHub.

Harold  
escribió:\> There are many ways to accomplish this, none of them are easy…

---

<div class="post-metadata">

**Author:** ![Arulalan](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/a/8c91f0/32.png) [@Arulalan](https://www.ruby-forum.com/u/Arulalan)\
**Post date:** [March 24, 2009, 8:21pm UTC](https://www.ruby-forum.com/t/ocr-project-in-gsoc/163767/3 "2009-03-24T20:21:03Z")

</div>

There are many ways to accomplish this, none of them are easy…

There’s ai4r’s backpropagation nueural nets implementation, with a  
simple OCR example at [http://ai4r.rubyforge.org/neuralNetworks.html](http://ai4r.rubyforge.org/neuralNetworks.html)

There’s also gnu Ocrad, which I’ve never used:  
[Ocrad - GNU Project - Free Software Foundation (FSF)](http://www.gnu.org/software/ocrad/),  
and just found [http://gtamilocr.sourceforge.net/](http://gtamilocr.sourceforge.net/) which does OCR for  
Tamil characters as well.

I’d be glad to hear other suggestions…