A Hand gesture recognition system based on difference image entropy

In this paper, we propose a real-time hand gesture recognition system based on the difference image entropy obtained using a stereo camera. In existing systems, hand detection has been primarily conducted in a constrained environment. However in this system, we implement a recognition system for incoming hand images in real-time. In the detection step, we implement a depth map using the SAD method based on right-left images acquired using a stereo camera. This system perceives the foreground object and performs hand detection. The hand gesture recognition system uses the difference image entropy of the input image and the average image. To evaluate the performance of the proposed technology, we implemented a recognition experiment using a hand gesture database (size 240). The experiment results show that the proposed method has an average recognition rate of 85%.