Communication Intelligence Corporation (CIC) proposes research on cursive handwriting recognition as a significant step towards developing more effective and ''user-friendly'' interfaces for man-to-machine communication. A promising new approach to user interfaces is based on recognition of handwriting and gesture. The goal is to provide the means by which common handwriting skills can be used to communicate with computers for a variety of applications such as word processing, interactive graphics, editing, etc. This has the potential to open up the use of computers to many people who find the conventional keyboard to be unnatural or intimidating, and to increase the efficiency of many applications for which handwriting and gesture are a more natural form of input. Before such interfaces can become viable commercially, however, advances in the state of the art in handwriting recognition are required. Current systems which require printing in boxes are too restrictive and unnatural for many applications CIC will evaluate a promising approach to achieving high performance cursive writing recognition based on a two stage precess involving: (1) a dynamic programming algorithm; and (2) a rulesbased discrimination techniques.Anticipated Results and
Potential Commercial Applications: It is expected that the proposed work will be incorporated into handwriting/gesture interfaces for such applications as CAD/CAM and desktop publishing. More generally, the importance of the proposed work is that it could lead to ''user-friendly" interfaces that could open up computer use to people who find the conventional keyboard unnatural or intimidating, and to interfaces that improve the efficiency of current applications.
