Application of gesture recognition based on simulated annealing BP neural network

In this paper, an algorithm of gesture recognition based on simulated annealing BP neural network is presented. Firstly, this new algorithm extracts the edge outline by skin color division and recognition feature of the distance between center and edge of binary gesture image. Secondly, it combines simulated annealing with BP neural network, which has both the learning ability and robustness of the neural network and the global optimization of simulated annealing, avoids the slow convergence and prevents it from falling into local minimum. The results of experiments show that the algorithm can greatly improve efficiency and accuracy of gesture recognition.