A Method Research of Hand Gesture Recognition Based on Hand Shape in Complex Background

Hand gesture recognition has become the hotspot in intelligent human-computer interaction. On the basis of the existing technology of hand gesture recognition, this paper presents a hand gesture recognition method based on the features of the hand in complex background. Use face recognition technology to determine the color threshold dynamically. It eliminates other interference region with the hands of geometric features such as the contour length and the area. Then use convex hull to detect fingertips and concave points fingers. Finally the procedure of recognition hand gestures is completed with the numbers of fingertips and concave points. The experimental results show that the proposed method is robust, and can recognize non-arm and arm gestures, and is able to adapt to the different environments.