Hand gesture recognition is very significant for human-computer interaction. In this work, we present a novel real-time method for hand gesture recognition. In our framework, the hand region is extracted from the background with Region of Interest.
The technique is implemented by reading a frame as an image then extracting only hand. Then, it is converting to black and white image. After that, a number is assigned to a gesture by counting number of flip as white to black from left to right on an intelligently selected path to be scanned on. We use the concept of Covex Hulls and Contours.