Monday, November 24, 2008

BanglaOCR V 0.5 (for Linux users) | Released

This is the announcement about the release of BanglaOCR V 0.5 (for Linux users). The release information is provided below:

Perquisites:
1. Tesseract OCR
2. Tidy
3. Java Runtime Environment (v 1.6)
4. Font : SolaimanLipi

User's Manual:
The manual is available with the package (BanglaOCR User Guide.pdf).

Feedback:
Any feedback about the application is highly appreciable. At present you can post feedback on the following email address: bangla.ocr@gmail.com

Test Images:
A set of test images are also available in the same site where the OCR setup package is available. So, please feel free to download them and test. The images were randomly selected to test the application and obviously not based on the best performance.

Download Link:
BanglaOCR V 0.5 (Linux Users)
Test Images

2 comments:

জয়ন্ত said...

Dear Hasnat,

Again I congratulate you for release for both OS ( Linux and windows). How may I help you regarding creating training data (total amount of training data units will be around 3200) or dictionary file? If you guide me, I can help you. As you know that your first concentration should be increase the recognition rate. I don’t know training data will be helpful for this or not. I know that it is just released version 0.6. and there have been lots of development to remain and I know that you should done it like at last nearest to Abbyy.

Md. Abul Hasnat said...

Dear Jayanta da,
thank you once again. I am planning to develop another application which will help us to prepare the training data in a sophisticated way. Right now its quite manual process to create the data and also painful to identify the errors in the training data. Considering this reason I am planning to develop a trainer so that anyone can prepare training data with a little bit knowledge about his own script.

I am highly appreciating all of your comments. Considering your eagerness I am highly encouraged to move forward on developing other required applications that might be helpful to generate training data. I am working on developing a "Text to Image Converter" so that it can help us to generate training data.

Keep in touch.