Gesture learning and recognition based on the Chebyshev polynomial neural network
Yang Zhiqi · 2016 IEEE Information Technology, Networking, Electronic and Automation Control Conference · 2016
Back propagation (BP) neural network and its improved algorithms have been widely used in gesture recognition, in order to improve the learning efficiency of the neural network for these algorithms, the author proposed an improved neural network algorithm based on Chebyshev polynomial, and applied the improved algorithm to the gesture recognition. The basic idea of this algorithm is: using Chebyshev polynomial as the activation function of the neural network, using the direct method based on the pseudo-inverse to obtain the weights of neural network; making reasonable definition and extraction of gesture samples; finally, using the improved neural network for dynamic gesture recognition. Experimental results show that the average training time of the new algorithm is 0.61 seconds, the correct recognition rate reaches 96.1%, so the efficiency of the new algorithm is much better than the average BP neural network and its ordinary improved algorithms.