PROJECT TITLE :

Vision-Based Remote Control System by Motion Detection and Open Finger Counting

ABSTRACT:

We describe a computer vision-based universal remote control system in this study. The procedure consists of two steps of visual evidence detection. Waving hands demanding control commands are detected using motion and skin color information. The camera is operated to zoom in on the local region of the hand in response to the request. The shape analysis on the segmented hand region image counts the number of open fingers. When a preset series of gesture state transitions is produced, a control command is given. The form of open fingers demonstrates strong properties for determining acceptable gesture states, according to experimental results. The application of stable features on successive frames produces reliable and precise results independent of operating conditions.