Real-time numeric character recognition system based on finger movements

In the field of computer vision, numeric character recognition plays an important role for various applications such as in Human Computer Interaction system. A numeric character recognition system is advantageous in a Human Computer Interaction system as it enables us to interact with a computer without a keyboard or other input devices. In this paper a simple numeric character recognition algorithm has been proposed where colored fingertip has been used for the purpose of tracking. Numeric characters are detected by the finger movements made during the handwriting of the characters. For each frame the tip position is compared to the previous tip position to decide whether the movement is a vertical or horizontal movement. Vertical movements can take place in upward or downward direction and horizontal movement can take place in rightward or leftward direction. This variation of fingertip movements gives rise to the classification criteria for the numeric characters. This method has shown promising result in real time environment.