Symbol Segmentation and Recognition for Understanding Handwritten Mathematical Expressions
Hanspeter Winkler, M. Lang · 1997
This paper is focused on the symbol segmentation and recognition problem within on-line sampled handwritten expressions, the first stage of an overall system for understanding arithmetic formulas. Within our system a statistical approach is used tolerating ambiguities within the single decision stages and resolving them either automatically by additional knowledge acquired during the following processing stages or by interaction with the user. At this state the interaction is done by displaying next to the recognition result, the most probable symbol sequence corresponding to the handwritten input, some recognition alternatives for selection by the user. 1