Hand gesture recognition is very significant for human-computer interaction. In this work, we present a novel real-time method for hand gesture recognition. The proposed system is vision based, which uses machine learning techniques and inputs from a computer webcam. Vision based gesture recognition tracking and gesture recognition in our framework, the hand region is extracted from the background with the background subtraction method. Finally, a rule classifier is applied to predict the labels of hand gestures. Our method shows better performance than a state-of-art method on another data set of hand gestures.