Detecting centroid for hand gesture recognition using morphological computations

With the evolution of computing technologies the current user computer interaction devices like mouse, keyboard, joysticks, pen etc. are getting replaced by touch screen and hand gesture recognition based user interfaces etc. Hand Gesture Recognition Systems are gaining importance as they provide a new ways to extract meaningful information from the gesture. The meaningful information is utilized to solve various problems in virtual reality, playing games, perform analysis of sign language etc. When using hand gestures to develop new user computer interaction method it is necessary to identify the correct gesture from the hand gesture image. In this paper using morphological operations we prepare the hand gesture image which can be used to recognize the correct hand gesture. The method is implemented to detect the centroid of hand gesture by removing the unwanted objects, noise and black holes if any in a static true color image whose background is uniform.