摘要
手写数字识别在多个领域具有广泛应用。在实际应用中,手写数字识别的准确性至关重要。为获得最佳的手写数字识别模型,文章提出一种用于手写数字识别的卷积神经网络模型,通过实验比较支持向量机、多层感知器和卷积神经网络模型在MNIST数据集手写数字识别中的识别准确性及其执行时间。实验结果表明,本文提出的卷积神经网络模型,在手写体数字数据集上的识别正确率可达99.4%。
Handwritten numeral recognition is widely used in many fields. In practical application, the accuracy of handwritten numeral recognition is very important. In order to obtain the best handwritten numeral recognition model, a convolution neural network model using handwritten numeral is proposed, and the recognition accuracy and execution time of support vector machine, multilayer perceptron and convolution neural network model in handwritten numeral recognition of MNIST data set are compared through experiments. The experimental results show that the recognition accuracy of the convolution neural network model proposed in this paper can reach 99.4%.
作者
张园
王书旺
马永兵
Zhang Yuan;Wang Shuwang;Ma Yongbing(School of Electronic Information,Nanjing Vocational College of Information Technology,Nanjing 210023,China)
出处
《信息化研究》
2021年第5期60-64,共5页
INFORMATIZATION RESEARCH
基金
南京信息职业技术学院自然科研基金项目(YK20180101)。
关键词
手写数字识别
MNIST数据集
支持向量机
多层感知器
卷积神经网络
handwritten numeral recognition
MNIST dataset
support vector machine
multilayer perceptron
convolutional neural network
