Identification Method of Digits for Expanding Touchpad Input

A method is presented for identifying the digits, i.e., the thumb and/or finger(s), that touch a touchpad as a means to expand the input vocabulary available on a touchpad. It will enable application designers to assign different commands to touch gestures performed with the same number of digits and the same movement but with different digits. No additional sensors are required for identification; instead the digits are identified on the basis of machine learning using only data acquired from a mutual-capacitance touchpad as learning data. Experimental results revealed an average identification accuracy of 86.3%.