基于CNN卷积神经网络的手势识别系统

Gesture Recognition System Based on CNN Convolutional Neural Network

与普通的神经网络非常相似,CNN卷积神经网络也由具有可学习的权重和偏置常量的神经元组成.每个神经元都接收一些输入并做一些点积计算,输出是每个分类的分数,普通神经网络里的一些计算技巧依旧适用.该文介绍了基于CNN卷积神经网络的手势识别系统,首先对不同的手势图片进行采集,将采集结果作为训练集和测试集,系统将会对其进行识别.

Similar to ordinary neural networks,CNN convolutional neural networks are also composed of neurons with learnable weights and bias constants.Each neuron receives some input and does some point product calculation.The output is the score of each classifica tion.Some calculation techniques in the general neural network are still applicable.In this paper,a gesture recognition system Based on CNN convolution neural network is introduced.Firstly,different gesture images are collected,and the collected results are used as train ing set and test set,which will be recognized by the system.