基于卷积神经网络的手势识别方法

A Method for Gesture Recognition Based on Convolutional Neural Network

手势识别是人机交互研究领域的一个重要分支,与人工提取图像特征的传统手势识别方法不同.设计一种基于卷积神经网络的手势识别方法,该方法使用包含3个卷积层、3个池化层、1个全连接层和1个Softmax回归层的卷积神经网络,自动地对手势样本进行特征提取、分类和迭代训练,提升了分类的准确性.实验结果表明,训练出的模型在测试集上的识别准确率可达到98.50%,并具有较强的鲁棒性.

Gesture recognition is an important branch of human-computer interaction research.A method for gesture recognition based on convolutional neural network is designed in this paper.The convolutional neural network consists of three convolutional layers,three pooling layers,one full-connection layer and one Softmax regression layer.This method is used to extract and classify gesture samples automatically by convolution neural network,and iterative training is carried out to improve the accuracy of classification.The experimental results show that the recognition accuracy of the proposed model is 98.50%on the test set and it has strong robustness.