P‐79: A Study of Prediction Model for Active Pen Coordinate Based on Machine Learning Algorithm

Abstract

Among digital pen technologies, Active Pen has the advantage of no additional price increase in functional implementation because it commonly uses capacitance touch sensors. However, since sensors designed for finger recognition purposes are used, coordinate accuracy is relatively low compared to digital pen technology that uses dedicated sensors. Therefore, in order to overcome these limitations, this paper studies the Active Pen position prediction algorithm by applying machine learning techniques. The R2 of the model derived in this study was about 0.999, and the RMSE of the predicted position was calculated to be about 10um.