Handwriting recognizer WritRecogn allows users to input characters used in Chinese, Japanese and Korean with a mouse or stylus.

WritRecogn recognizes handwritten characters for Chinese, Japanese, and Korean (CJK) and will interface to input method framework such as SCIM. Typical approaches require either a huge set of complex character recognition rules built by professional linguists and developers, or character examples diligently drawn by end-users. The former approach is not suitable for community-developed project because professional linguists might not be available; while the latter approach does not work for CJK users as nobody will draw tens of thousand characters before the recognizer can be practically used.

WritRecogn addresses these problems by automatically generates recognition rules character formation information (later referred as "textbooks") which are easy enough for amateurs to comprehend and manipulate.

The main program which provides the GUI is WritRecogn and there is a command line character data maintenance program WritRecogn-manager. Other features include: Stroke editor: users can input new characters for the recognizer to learn.



Version 0.1.9 Released: 2008-04-24

Basically it is refined version of 0.1.8. However, there is something new:

Version 0.1.8 Released: 2008-03-26

It's a preview version of 0.2, though the interface is slightly change, the program is almost completed rewritten. Mainly for GOBize the main data structures, such as Radical, abstract characters, and raw strokes.

The reason for it is only 0.1.8 is because the XML file format does yet support the Radical bounding boxes. and Tomoe and Wubi import functions have not been converted yet.

Further details is at Release Note.