How a Recognizer Works

There are two techniques for recognizing handwriting, called bitmap and vector recognition.

Bitmap recognition attempts to match an ink image with a record of known character images. The bitmap recognizer sees the ink data as a stencil pattern of points that it can compare to a library of patterns, searching for the closest match. This technique, employed by optical character recognizers (OCRs), works well for patterns limited to a few styles and sizes.

In contrast, vector recognition sees the ink as lines rather than points. The method considers characteristics of the lines collected as the pen moves. These characteristics include sequence, curvature, direction, and so forth. Given the wide varieties and styles of handwriting, vector recognition works best for deciphering pen input. The Pen API does not mandate which method a recognizer employs, but is designed to facilitate vector rather than bitmap recognition.

Software for developers
Delphi Components
.Net Components
Software for Android Developers
More information resources
Unix Manual Pages
Delphi Examples