ONLINE HANDWRITING RECOGNITION.

This article presents information related to online handwriting recognition, that has been a research challenge for almost 40 years. Contrary to offline techniques, which deal with the automatic processing of the image of handwritten words as collected using a camera or a scanner, online approaches aim at recognizing words as they are written, using a digitizer or an instrumented pen to capture pen-tip position as a function of time. The major difference between online and offline data is that the time sequence of movements for online data is directly available from the acquired signals. This allows also the computation of other signals, like velocity signals, from the original data.