Hand gesture recognition using skeleton of hand and distance based metric

Abstract

In this paper we are mainly concerns on the image processing and computer vision concepts for interpretation of gestures .By using gestures we can convey instructions to the machine (computer) or commands to a robots .This is known as Human machine interaction (Human computer interaction (HCI).Hand gestures are an ideal way of exchanging information between human and computer, robots, or any other device. In this paper we are calculating skeleton of the hand by using distance transformation technique and are using for recognition instead of the entire hand, because of its robust nature against translation, rotation and scaling. Skeleton is computed for each and every hand posture in the entire hand motion and superimposed on a single image called as Dynamic Signature of the particular gesture type. Gesture is recognized by using the Image Euclidean distance measure by comparing the current Dynamic Signature of the particular gesture with the gesture Alphabet set. © 2011 Springer-Verlag.