Hand-Gesture Recognition-Algorithm based on Finger Counting

Abstract

The concept of hand gesture recognition has been widely used in communication, artificial intelligence, and robotics. The most contributing reason for the emerging gesture recognition is that they can create a simple communication path between human and computer called HCI (Human-Computer Interaction). Therefore, a hand gesture recognition algorithm was developed for fourteen hand gestures based on finger counting. The algorithm counts fingers and recognizes gesture based on the maximum distance between the fingers detected. The algorithm divided into four main parts: image acquisition, pre-processing, finger detection, and gesture recognition. The experimental results show that the algorithm can count fingers accurately and recognize 10 gestures (associated with 1, 2, 3 and 5 fingers) with good performance (70 to 100 percent of successful detection) and 4 gestures (associated with 4 fingers) with average performance (50 to 70 percent of successful detection). Additionally, the algorithm was tested under variation of the scene and dynamic parameters, to understand its performance further.