Digital Image Processing Based Hand Gesture Recognition Robot for Multiple Applications

Abstract: In early days, for operation of robots a person should be present beside it. So it is necessary to design a robot which can be operated without a person being physically present. This paper is about operating a robot which is not near us and can be operated from a station; this is possible and is executed through a technique called IMAGE PROCESSING. In order to ease a feasible solution for user friendly interface, we can give commands to a wireless robot using hand gestures. Hand Gesture is frequently used in day to day life. The gesture recognition system consists of three stages image acquisition, feature extraction and decision algorithm. Once a gesture is recognized a command signal is generated and sent to micro controller. These command signals are then sent to the robot to operate it in a particular direction. One of the applications of the robot is to pick and place things where human intervention is not possible. The pick and place arm is a microcontroller based mechatronics system that picks the object from a source location and places it at a desired location, this action is controlled using a keypad switch. Further, the robot captures the video signals with its RF camera at 2.4GHz (mounted on the robot). The live video of the robot is received at the base station (computer). It is processed in the MAT Lab software and the video is processed in DSP code simultaneously.

Keywords: Digital Image Processing, Hand Gesture Recognition, Robot, MATLAB.