NHocr is a command line OCR (Optical Character Recognition) program for Japanese language, etc. It has been designed to recognize machine-printed Japanese characters and some ASCII characters/symbols in an image. NHocr is probably the first Open Source Japanese OCR software (offline, machine-printed), except some experimental, partial codes open to academic communities.
The main repository, originally at Google Code, has been migrated to here. Older versions can be found at Google Code.
You can test-drive NHocr using the following:
* Japanese character recognition WeOCR service
* Capture2Text
NHocr ver 0.22 source code distribution
* Some code fixes have been made.
* A part of the image manipulation library O2-tools has been included in the source tree.
* Vertical writing has been supported. However, the accompanying dictionaries in this release lack vertical fonts, especially in symbols. Many characters are compatible and could be recognized well also in vertical documents.
The dictionaries with vertical fonts will be available in future releases.
