In-Tool Motion Sensing for Evaluation of Violin Performance

Abstract

A system that is able to measure motions is necessary for preserving the skills of experts. A previously reported study proposed a system that precisely measures and reproduces motions. However, this type of system must physically interact with the subject whose motions are measured. Such interaction can be troublesome when the measured motions require precise control of force or position or even a high degree of freedom. Therefore, this study proposes a novel method for sensing motions without interference and applies the approach to measure the skill of playing a violin. Using the data measured by the proposed system, the skills of an experienced violin player and a novice violin player are compared. This study attempts to visualize the differences between the two types of players by applying wavelet transform on the measured data and analyzing them in the frequency domain.