Design of children's game machine system based on gesture recognition
Gesture recognition technology is an important part of human-computer interaction.At the present stage,gesture rec-ognition technology has some problems such as inaccurate location of key points and poor effect of gesture recognition,so it is difficult to make a breakthrough in the usable field.Based on this,the method of using improved BP neural network is proposed to improve image key point recognition.In addition,through the gesture recognition module based on ECPM and the gesture pre-module of LSTM,we jointly build a children's game machine based on gesture recognition.The performance of the algorithm and the game ma-chine is verified by experiments.The results show that the average recognition rate of gesture is 97.89％and the average prediction accuracy is 75.42％.This study provides a reference for the improvement of gesture recognition and a scheme for the design of chil-dren's game machines.
万方数据
维普
