Digit recognition from pressure sensor data using Euler number and central moments

The work described in the paper deals with realization of a digit recognition system for the visually impaired persons. The input to the system is pressure sensor data obtained from a sensor pad with 256 capacitive pressure sensors. The objective of the system is to recognize correctly the 2-D shape of a digit which is placed on the above sensor pad. The problem is extremely challenging because the input data i) is very susceptible to noise and ii) may undergo rigid body transformation with respect to the reference library of the system. During the first step of the recognition process, the Euler number is used to achieve a preliminary classification. In the second step, 4^th order central moment is applied to correctly recognize the input. A comparison of the performance of the proposed method with some other existing approaches indicates our recognition results to be highly promising.