A Vision-Based Hand Gesture Recognition System: Development and Modelling

Gesture recognition systems are gaining popularity nowadays because they can interface with machines via human–computer interaction. The development of such systems helps to develop a more natural communication between man and machine. These setups made humans directly instruct the machine as per their needs. A vision-based gesture recognition setup has been developed and modelled in the present study. The gesture recognition system has been developed by a Linux processing device with a camera and an output control device. MediaPipe Hands model has been used for gesture recognition. Programming for the used model has been done using Python programming language. The Arduino UNO has also been programmed to convert the gestures into desired outputs. The developed vision-based gesture recognition system was found to work effectively for human hand gestures.