A Data Format defines the linguistic, syntactic, and semantic constraints of the handwritten information to be recognized by the recognition engine; it could be described as the expected type of input. Your handwriting context's Data Format may be textual, and could be recognized with the help of a lexicon, or it could be more specific, such as a date or time, in which case it could be defined in the form of a regular expression. MyScript Builder offers language-specific Data Formats to assist in the recognition.
A series of strokes following the trajectory of the user's handwriting, entered using a pen or pointing device through pen interfaces such as a PDA touch screen, TabletPC touch screen, Smartphone touch screen, Digital Pen, Graphics Tablet, Interactive Whiteboard, and so on.
Isolated Characters: Handwriting style in which characters are separated individually, in boxes for example.
Hand printed writing: Handwriting style in which characters are not necessarily separated physically (e.g in boxes) but with which each character must be fully formed (including any diacritical marks such as accents) before starting the next. A distinct pen lift must occur between the two characters.
Cursive handwriting: Handwriting style, also called "joined-up" or "running writing" or "natural writing", in which letters may or may not be joined as you write.
A method for recognizing hand written text. Data input can be offline (image from a scanner) but is most commonly online from a pen or pointing device. The text is analyzed to identify characters or digits and this analysis is then translated into a character code system such as ASCII. ICR is often used as a synonym for handwriting recognition.
A lexicon is a vocabulary list: it does contain words, typically, but it may also contain groups of words such as proper names, brand names, trademarks and other lexical expressions (which may include separators) that only make sense when they are kept together.
A digital pen or pointing device is an input peripheral that is able to capture digital ink, be that the user's handwriting, drawings, scribbles or other handwritten material.
An expression defined using operators that specifies what characters to expect in an input unit and in which order. These are used to define certain lexical units that you wish to recognize, such as dates, prices, etc.
Resource files are files used by the recognition engine to assist it in the recognition process, for example, the files that identify handwriting styles are resources, as are files used to recognize data formats, based on lexicons, or regular expressions. This knowledge is compiled into a file format, ready to be attached to a recognizer.
A process by which the recognition engine breaks digital ink into input ranges. This occurs at a text, word and character level.