What’s up folks!? I hope everybody is enjoying some good holidays or getting prepared to do so soon.
As for me, I am currently in Portugal to spend Christmas with my family and I have just released a new version of OCRFeeder, its 0.5 version. This will be the last version released in 2009.
So, what’s new in version 0.5?
Hongde Liu (刘洪德), a Chinese user, was kind enough to provide me with the Simplified Chinese translation of OCRFeeder.
This will hopefully allow more GNOME users from China to use what is the most complete Open Source OCR application for GNOME.
The engines Tesseract, Ocrad and GOCR are now automatically detected from the system when no engines are configured.
Regarding the OCR engines, some verifications were also introduced to prevent errors when no engines are used.
The bounding boxes are now restricted to be dragged inside the image’s limits, that is, it will no longer allow a box to be dragged beyond the image’s borders.
For users interested in the development, the Makefile now features the generatepot and compilemessages commands which generate the application’s pot file and compile a po file given its language code, respectively.
I plan to replace all minidom code by ElementTree, so, since this version, the feeder module is uses ElementTree.
As usual, get OCRFeeder’s source from Gitorious or download a Debian package.
Thanks, cool stuff! In addition to the Tesseract, also I’m using OCR Cuneiform. OCRFeeder can be made to work with Cuneiform? If so, please tell me how it can be configured.
OCRFeeder was written to work with whatever engines you can run from the command line.
I never used Cuneiform but let me know how do you usually use it and I’ll post on how to use it with OCRFeeder.
Cheers,