ECONOMICAL AND USER-FRIENDLY DESIGN OF VISION-BASED NATURAL-USER INTERFACE VIA DYNAMIC HAND GESTURES

A vision-based hand gesture recognition technology can bring a revolutionary and beneficial change in the field of human-machine interaction for elderly and special people living at home. Nevertheless, continuous detection and localization of the moving hand region in true-color images are strenuous tasks because the hand is a non-rigid object and occupies a small area in the whole frame. True-color images are also sensitive to light variation, camera-view, and background conditions. To ease the process of hand detection and tracking, researchers prefer advanced cameras equipped with costly sensors. This increases the overall cost of interfaces and also requires technical knowledge to operate them. The second issue in dynamic hand gesture recognition is the unpredictability of hand pose used by the user and the random behavior of the hand movement while performing the hand gesture. Therefore, the use of dynamic hand gestures in vision-based human-machine interaction is limited. The goal of this paper is to propose an economical, and user-friendly, technique for vision-based human-machine interaction via dynamic hand gestures that is user-friendly and affordable by all. The proposed technique can be integrated with any day-to-day machines, for example, washing machines, radio, fans, automated doors, etc.