基于改进Sigmoid卷积神经网络的手写体数字识别

doi:10.11896/jsjkx.211000179

计算机科学 ›› 2022, Vol. 49 ›› Issue (12): 244-249.doi: 10.11896/jsjkx.211000179

• 计算机图形学&多媒体 • 上一篇下一篇

基于改进Sigmoid卷积神经网络的手写体数字识别

樊继慧^1,2, 滕少华³, 金弘林²

1 圣保罗大学研究生院卡加延土格加劳3500
2 广州理工学院计算机科学与工程学院广州510540
3 广东工业大学计算机学院广州510006

收稿日期:2021-10-25 修回日期:2022-02-25 发布日期:2022-12-14
通讯作者: 樊继慧(2519639989@qq.com)
基金资助:
国家自然科学基金(61972102) ;广东省教育厅重大专项(2021ZDZX1070);广东省高等教育研究课题(22GQN37);校本研究项目(2021XBZ03)

Handwritten Numeral Recognition Based on Improved Sigmoid Convolutional Neural Network

FAN Ji-hui^1,2, TENG Shao-Hua³, JIN Hong-Lin²

1 Graduate School,St.Paul University Philippines,Tuguegarao,Cagayan 3500,Philippines
2 School of Computer Science and Engineering,Guangzhou Institute of Science and Technology,Guangzhou 510540,China
3 School of Computer Science and Engineering,Guangdong University of Technology,Guangzhou 510006,China

Received:2021-10-25 Revised:2022-02-25 Published:2022-12-14
About author:FAN Ji-hui,born in 1990,postgraduate,lecturer.Her main research interests include data analysis and mining.
Supported by:
National Natural Science Foundation of China(61972102),Major Special Projects of Guangdong Provincial Department of Education(2021ZDZX1070),Guangdong Higher Education Research Project(22GQN37) and School Based Research Project(2021XBZ03).

摘要/Abstract

摘要： 深度学习技术在数字识别领域有着普遍的应用。通过深度学习技术构造神经网络模型,运用不同的激活函数搭配不同的参数初始化策略,对MINIST手写数据集进行训练;构建分析模型,识别图像中的数字,将大数据量的图片降维成小数据量图片,同时保证能够有效保留图片特征;通过对图片数据的分析,加入特征转换过程,利用梯度下降优化器,搭建网络结构,将数据降维,有效地避免过拟合;利用交叉熵验证对模型进行编译和训练,对输出的分类结果进行进一步分析,在Sigmoid激活函数的输出层,通过K最近邻分类算法,设置KNN分类器,进一步提高了分类预测的准确率。MNIST数据集上的实验结果显示识别率为96.2%,在输出层引入K最近邻算法KNN(K-Nearst Neighbors)结合传统卷积神经网络(Convolutional Neural Network,CNN)的全连接层与softmax层,经交叉验证得到99.6%的识别率。

关键词: 数字识别, K最近邻算法, 深度学习, 卷积神经网络, 交叉熵

Abstract: Deep learning technology is widely used in the field of number recognition.It constructs neural network model through deep learning technology,nonlinear transformation activation function in neurons,different activation functions with different parameter initialization strategies,trains MINIST handwritten data set,constructs analysis model and recognizes numbers in images,reduce the dimension of a large amount of data into a small amount of data,and ensure the effective retention of image features.Through the analysis of image data,adding the feature conversion process,using the gradient descent optimizer to build a network structure and reduce the dimension of data,which can effectively avoid over fitting.Cross-entropy verification is used to compile and train the model,and the output classification results are further analyzed.Through the K-nearest neighbor classification algorithm,KNN classifier is set to further improve the accuracy of classification and prediction.Through MNIST data set experiment,the recognition rate is about 96.2%.The K-nearest neighbor algorithm(KNN) is introduced into the output layer,combined with the full connection layer and softmax layer of traditional convolutional neural network(CNN).After cross verification,the recognition rate is 99.6%.

Key words: Digital identification, K nearest neighbor algorithm, Deep learning, Convolutional neural network, Cross entropy

中图分类号:

TP391

樊继慧, 滕少华, 金弘林. 基于改进Sigmoid卷积神经网络的手写体数字识别[J]. 计算机科学, 2022, 49(12): 244-249. https://doi.org/10.11896/jsjkx.211000179

FAN Ji-hui, TENG Shao-Hua, JIN Hong-Lin. Handwritten Numeral Recognition Based on Improved Sigmoid Convolutional Neural Network[J]. Computer Science, 2022, 49(12): 244-249. https://doi.org/10.11896/jsjkx.211000179

参考文献

[1]YIN Z Y,XU G Y,ZHANG F Q,et al.Design and implementation of convolutional neural network unit for zynq platform [J/OL].Small microcomputer system:1-6.http://kns.cnki.net/kcms/detail/21.1106.TP.20210413.1558.012.html.
[2]WANG A L,XUE D,WU H B,et al.Handwritten digit recognition based on conditional generation countermeasure network [J].Liquid Crystal and Display,2020,35(12):1284-1290.
[3]MA Y X,GUO X J.Recognition of handwritten digits based on K-nearest neighbor algorithm[J].Journal of Jilin Jianzhu University,2021,8(6):84-86.
[4]ZHU Y,CHEN S P.Text classification model of nearest neighbor attention and convolutional neural network [J].Small Microcomputer System,2020,41(2):375-380.
[5]DUAN L L,ZHANG L.A handwritten numeral recognitionmethod based on improved convolutional neural network [J].Modern Electronic Technology,2021,44(18):107-110.
[6]WANG M,LI D X.Handwritten numeral recognition based on improved vgg-16 and naive Bayes [J].Modern Electronic Technology,2020,43(12):176-181,186.
[7]WANG J R,MA X,DUAN G L,et al.Handwritten numeral recognition under the background of edge intelligence [J].Computer Applications,2019,39(12):3548-3555.
[8]SONG X R,WU X,GAO S,et al.Simulation Research on handwritten digit recognition based on deep neural network [J].Science,Technology and Engineering,2019,19(5):193-196.
[9]ZENG W X,MENG Q L,GUO Z K.Research on handwritten digit recognition based on deep convolution self coding neural network [J].Computer Application Research,2020,37(4):1239-1243.
[10]CHEN Y,LI Y Y,YU L,et al.Handwritten digit recognition system based on convolutional neural network [J].Microelectronics and Computer,2018,35(2):71-74.
[11]XIE D Y.LI L H,MIAO C S.Handwritten digit recognition based on improved alexnet convolutional neural network[J].Journal of Hebei University of Engineering,2021,38(4):102-106.
[12]DU X.Research and implementation of handwritten digit recognition car based on Wireless Bluetooth transmission [J].Electronic Testing,2022,36(6):8-10.
[13]HAN X,LIU Q,XU J,et al.Handwritten numeral recognition algorithm based on pseudo PCA [J].Computer Science,2018,45(S2):278-281,307.
[14]GAO Y P,HU Z Y.Improvement of handwritten numeral re-cognition model based on keras [J].Computing Technology and Automation,2021,40(2):164-169.
[15]ZHAO J L,HU L,YAN H,et al.Hyperspectral image classification method combining local binary mode and k-nearest neighbor algorithm [J].Journal of Infrared and Millimeter Wave,2021,40(3):400-412.
[16]LI S J,ZHANG K X,WANG S D,et al.Feature selection me-thod based on weighted k-nearest neighbor [J].Journal of Tianjin Normal University(Natural Science Edition),2020,40(2):63-67.
[17]YIN Y B,YANG W Z,YANG H T,et al.Improved KNN Text Classification Algorithm Based on search [J].Computer Engineering and Design,2018,39(9):2923-2928.
[18]HUANG W M,MO Y.Chinese spam message filtering based on text weighted KNN algorithm [J].Computer Engineering,2017,43(3):193-199.

相关文章 15

[1]	周乐员, 张剑华, 袁甜甜, 陈胜勇. 多层注意力机制融合的序列到序列中国连续手语识别和翻译 Sequence-to-Sequence Chinese Continuous Sign Language Recognition and Translation with Multi- layer Attention Mechanism Fusion 计算机科学, 2022, 49(9): 155-161. https://doi.org/10.11896/jsjkx.210800026
[2]	徐涌鑫, 赵俊峰, 王亚沙, 谢冰, 杨恺. 时序知识图谱表示学习 Temporal Knowledge Graph Representation Learning 计算机科学, 2022, 49(9): 162-171. https://doi.org/10.11896/jsjkx.220500204
[3]	饶志双, 贾真, 张凡, 李天瑞. 基于Key-Value关联记忆网络的知识图谱问答方法 Key-Value Relational Memory Networks for Question Answering over Knowledge Graph 计算机科学, 2022, 49(9): 202-207. https://doi.org/10.11896/jsjkx.220300277
[4]	汤凌韬, 王迪, 张鲁飞, 刘盛云. 基于安全多方计算和差分隐私的联邦学习方案 Federated Learning Scheme Based on Secure Multi-party Computation and Differential Privacy 计算机科学, 2022, 49(9): 297-305. https://doi.org/10.11896/jsjkx.210800108
[5]	陈泳全, 姜瑛. 基于卷积神经网络的APP用户行为分析方法 Analysis Method of APP User Behavior Based on Convolutional Neural Network 计算机科学, 2022, 49(8): 78-85. https://doi.org/10.11896/jsjkx.210700121
[6]	朱承璋, 黄嘉儿, 肖亚龙, 王晗, 邹北骥. 基于注意力机制的医学影像深度哈希检索算法 Deep Hash Retrieval Algorithm for Medical Images Based on Attention Mechanism 计算机科学, 2022, 49(8): 113-119. https://doi.org/10.11896/jsjkx.210700153
[7]	孙奇, 吉根林, 张杰. 基于非局部注意力生成对抗网络的视频异常事件检测方法 Non-local Attention Based Generative Adversarial Network for Video Abnormal Event Detection 计算机科学, 2022, 49(8): 172-177. https://doi.org/10.11896/jsjkx.210600061
[8]	檀莹莹, 王俊丽, 张超波. 基于图卷积神经网络的文本分类方法研究综述 Review of Text Classification Methods Based on Graph Convolutional Network 计算机科学, 2022, 49(8): 205-216. https://doi.org/10.11896/jsjkx.210800064
[9]	李宗民, 张玉鹏, 刘玉杰, 李华. 基于可变形图卷积的点云表征学习 Deformable Graph Convolutional Networks Based Point Cloud Representation Learning 计算机科学, 2022, 49(8): 273-278. https://doi.org/10.11896/jsjkx.210900023
[10]	王剑, 彭雨琦, 赵宇斐, 杨健. 基于深度学习的社交网络舆情信息抽取方法综述 Survey of Social Network Public Opinion Information Extraction Based on Deep Learning 计算机科学, 2022, 49(8): 279-293. https://doi.org/10.11896/jsjkx.220300099
[11]	郝志荣, 陈龙, 黄嘉成. 面向文本分类的类别区分式通用对抗攻击方法 Class Discriminative Universal Adversarial Attack for Text Classification 计算机科学, 2022, 49(8): 323-329. https://doi.org/10.11896/jsjkx.220200077
[12]	姜梦函, 李邵梅, 郑洪浩, 张建朋. 基于改进位置编码的谣言检测模型 Rumor Detection Model Based on Improved Position Embedding 计算机科学, 2022, 49(8): 330-335. https://doi.org/10.11896/jsjkx.210600046
[13]	胡艳羽, 赵龙, 董祥军. 一种用于癌症分类的两阶段深度特征选择提取算法 Two-stage Deep Feature Selection Extraction Algorithm for Cancer Classification 计算机科学, 2022, 49(7): 73-78. https://doi.org/10.11896/jsjkx.210500092
[14]	张颖涛, 张杰, 张睿, 张文强. 全局信息引导的真实图像风格迁移 Photorealistic Style Transfer Guided by Global Information 计算机科学, 2022, 49(7): 100-105. https://doi.org/10.11896/jsjkx.210600036
[15]	戴朝霞, 李锦欣, 张向东, 徐旭, 梅林, 张亮. 基于DNGAN的磁共振图像超分辨率重建算法 Super-resolution Reconstruction of MRI Based on DNGAN 计算机科学, 2022, 49(7): 113-119. https://doi.org/10.11896/jsjkx.210600105

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

基于改进Sigmoid卷积神经网络的手写体数字识别

Handwritten Numeral Recognition Based on Improved Sigmoid Convolutional Neural Network

PDF (PC)

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

Metrics

本文评价

推荐阅读 0