A COMPUTATIONAL-PROCEDURE FOR MOVEMENT ANALYSIS IN HANDWRITING

Validity of handwriting movement data recorded with digitizing tablets is poorly reflected by manufacturer's specifications. Actual errors o f positional data are in fact in a critical range, in particular when time derivatives are calculated. Because differentiation magnifies err ors in the displacement data, data smoothing becomes crucial. A new me thod for smoothing and differentiating noisy handwriting movement data is proposed. Non-parametric estimation of regression functions using kernel estimates generally offers simple application and extremely fas t calculation. Assessing simulation data the efficiency of the procedu re was investigated and compared with butterworth filters and finite i mpulse response (FIR) filters. Kernel estimates show a slightly elevat ed bias, but strongly reduced residual variance in the velocity and ac celeration signals. The overall smoothing behaviour of kernel estimate s is better than butterworth filters and very similar to FIR filters, if their transition band is chosen appropriately.