Combination of Convolutional Neural Network Architecture and its Learning Method for Rotation-Invariant Handwritten Digit Recognition

被引：4

作者：

Urazoe, Kazuya ^{[1
]}

Kuroki, Nobutaka ^{[1
]}

Hirose, Tetsuya ^{[2
]}

Numa, Masahiro ^{[1
]}

机构：

[1] Kobe Univ, Dept Elect & Elect Engn, Nada Ku, 1-1 Rokkodai Cho, Kobe, Hyogo 6578501, Japan

[2] Osaka Univ, Div Elect Elect & Informat Engn, 2-1 Yamadaoka, Suita, Osaka 5650871, Japan

来源：

IEEJ TRANSACTIONS ON ELECTRICAL AND ELECTRONIC ENGINEERING | 2021年 / 16卷 / 01期

关键词：

handwritten digit recognition; convolutional neural network; rotation invariance; data augmentation; MNIST;

D O I：

10.1002/tee.23278

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

This letter presents several combinations of a convolutional neural network (CNN) and its learning method for rotation-invariant digit recognition. Rotation data augmentation is widely used for improving rotation invariance. Data augmentation commonly assigns the same label to all augmented images of the same source. However, this learning method causes some collisions between original and rotated digits. Thus, this letter presents three types of rotation-invariance learning methods and applies them to five popular CNN architectures. Experimental results indicate that multi-task learning on ResNet-50 is the best combination. (c) 2020 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC.

引用

页码：161 / 163

页数：3

共 50 条

[1] A Convolutional Neural Network for Handwritten Digit Recognition
Guevara Neri, Maria Cristina
Vergara Villegas, Osslan Osiris
Cruz Sanchez, Vianey Guadalupe
Nandayapa, Manuel
Sossa Azuela, Juan Humberto
INTERNATIONAL JOURNAL OF COMBINATORIAL OPTIMIZATION PROBLEMS AND INFORMATICS, 2020, 11 (01): : 97 - 105
[2] A Multi-Scale Convolutional Neural Network for Rotation-Invariant Recognition
Hong, Tzung-Pei
Hu, Ming-Jhe
Yin, Tang-Kai
Wang, Shyue-Liang
ELECTRONICS, 2022, 11 (04)
[3] Handwritten Digit Recognition Based on Convolutional Neural Network
Zhang, Chao
Zhou, Zhiyao
Lin, Lan
2020 CHINESE AUTOMATION CONGRESS (CAC 2020), 2020, : 7384 - 7388
[4] Handwritten Digit String Recognition using Convolutional Neural Network
Zhan, Hongjian
Lyu, Shujing
Lu, Yue
2018 24TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION (ICPR), 2018, : 3729 - 3734
[5] Bangla Handwritten Digit Recognition Using Convolutional Neural Network
Rabby, A. K. M. Shahariar Azad
Abujar, Sheikh
Haque, Sadeka
Hossain, Syed Akhter
EMERGING TECHNOLOGIES IN DATA MINING AND INFORMATION SECURITY, IEMIS 2018, VOL 1, 2019, 755 : 111 - 122
[6] A ROTATION-INVARIANT CONVOLUTIONAL NEURAL NETWORK FOR IMAGE ENHANCEMENT FORENSICS
Chen, Yifang
Lyu, Zixian
Kang, Xiangui
Wang, Z. Jane
2018 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP), 2018, : 2111 - 2115
[7] Rotation-Invariant Feature Learning via Convolutional Neural Network With Cyclic Polar Coordinates Convolutional Layer
Mei, Shaohui
Jiang, Ruoqiao
Ma, Mingyang
Song, Chao
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2023, 61
[8] Morphological Convolutional Neural Network Architecture for Digit Recognition
Mellouli, Dorra
Hamdani, Tarek M.
Sanchez-Medina, Javier J.
Ben Ayed, Mounir
Mimi, Adel M.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2019, 30 (09) : 2876 - 2885
[9] RIC-CNN: Rotation-Invariant Coordinate Convolutional Neural Network
Mo, Hanlin
Zhao, Guoying
PATTERN RECOGNITION, 2024, 146
[10] Deep Convolutional Neural Network for Handwritten Bangla and English Digit Recognition
Akbar, Md Ali
Islam, Md Saiful
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON ELECTRONICS, COMMUNICATIONS AND INFORMATION TECHNOLOGY 2021 (ICECIT 2021), 2021,

← 1 2 3 4 5 →