Handwritten Character and Digit Recognition with Deep Convolutional Neural Networks: A Comparative Study

Chui En Mook; Chin Poo Lee; Kian Ming Lim; Jit Yan Lim

doi:10.1109/ICoICT58202.2023.10262721

Handwritten Character and Digit Recognition with Deep Convolutional Neural Networks: A Comparative Study

Chui En Mook, Chin Poo Lee, Kian Ming Lim, Jit Yan Lim

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research

8 Citations (Scopus)

Abstract

Handwritten character or digit recognition involves automatically classifying handwritten characters or digits from images. Previous studies focused on specific datasets and did not thoroughly compare different CNN architectures. This paper addresses these limitations by presenting a comparative study of six popular CNN architectures (VGG16, Xception, ResNet152V2, InceptionResNetV2, MobileNetV2, and DenseNet201) on three diverse datasets: English Handwritten Characters, Handwritten Digits, and MNIST. The experimental results demonstrate that the InceptionResNetV2 model with data augmentation achieves the highest accuracy across all datasets, with accuracies of 93.26%, 97.16%, and 99.71% on the English Handwritten Characters, Handwritten Digits, and MNIST datasets, respectively.

Original language	English
Title of host publication	11th International Conference on Information and Communication Technology, ICoICT 2023
Editors	Lee-Ying Chong, Tee Connie, Dawam Dwi Jatmiko Suwawi, Joon Liang Tan
Place of Publication	Piscataway NJ USA
Publisher	IEEE, Institute of Electrical and Electronics Engineers
Pages	137-141
Number of pages	5
ISBN (Electronic)	9798350321982
ISBN (Print)	9798350333039
DOIs	https://doi.org/10.1109/ICoICT58202.2023.10262721
Publication status	Published - 2023
Externally published	Yes
Event	International Conference on Information and Communication Technology 2023 - Melaka, Malaysia Duration: 23 Aug 2023 → 24 Aug 2023 Conference number: 11th https://ieeexplore.ieee.org/xpl/conhome/10262402/proceeding (Proceedings) https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=&ved=2ahUKEwiKrtbNp-GEAxUbs1YBHZlDAVkQFnoECA8QAQ&url=https%3A%2F%2Fwww.icoict.org%2F2023-icoict%2F&usg=AOvVaw2TRUvwZYzGEHUrk65UNNeI&opi=89978449 (Website)

Conference

Conference	International Conference on Information and Communication Technology 2023
Abbreviated title	ICoICT 2023
Country/Territory	Malaysia
City	Melaka
Period	23/08/23 → 24/08/23
Internet address	https://ieeexplore.ieee.org/xpl/conhome/10262402/proceeding (Proceedings) https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=&ved=2ahUKEwiKrtbNp-GEAxUbs1YBHZlDAVkQFnoECA8QAQ&url=https%3A%2F%2Fwww.icoict.org%2F2023-icoict%2F&usg=AOvVaw2TRUvwZYzGEHUrk65UNNeI&opi=89978449 (Website)

Keywords

Convolution Neural Network
Data Augmentation
Handwritten Character Recognition
Handwritten Digit Recognition

Access to Document

10.1109/ICoICT58202.2023.10262721

Cite this

Mook, C. E., Poo Lee, C., Lim, K. M., & Yan Lim, J. (2023). Handwritten Character and Digit Recognition with Deep Convolutional Neural Networks: A Comparative Study. In L.-Y. Chong, T. Connie, D. Dwi Jatmiko Suwawi, & J. Liang Tan (Eds.), 11th International Conference on Information and Communication Technology, ICoICT 2023 (pp. 137-141). IEEE, Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/ICoICT58202.2023.10262721

Mook, Chui En ; Poo Lee, Chin ; Lim, Kian Ming et al. / Handwritten Character and Digit Recognition with Deep Convolutional Neural Networks : A Comparative Study. 11th International Conference on Information and Communication Technology, ICoICT 2023. editor / Lee-Ying Chong ; Tee Connie ; Dawam Dwi Jatmiko Suwawi ; Joon Liang Tan. Piscataway NJ USA : IEEE, Institute of Electrical and Electronics Engineers, 2023. pp. 137-141

@inproceedings{dfdcfb4641514bdc908f5db07a7669cb,

title = "Handwritten Character and Digit Recognition with Deep Convolutional Neural Networks: A Comparative Study",

abstract = "Handwritten character or digit recognition involves automatically classifying handwritten characters or digits from images. Previous studies focused on specific datasets and did not thoroughly compare different CNN architectures. This paper addresses these limitations by presenting a comparative study of six popular CNN architectures (VGG16, Xception, ResNet152V2, InceptionResNetV2, MobileNetV2, and DenseNet201) on three diverse datasets: English Handwritten Characters, Handwritten Digits, and MNIST. The experimental results demonstrate that the InceptionResNetV2 model with data augmentation achieves the highest accuracy across all datasets, with accuracies of 93.26\%, 97.16\%, and 99.71\% on the English Handwritten Characters, Handwritten Digits, and MNIST datasets, respectively.",

keywords = "Convolution Neural Network, Data Augmentation, Handwritten Character Recognition, Handwritten Digit Recognition",

author = "Mook, \{Chui En\} and \{Poo Lee\}, Chin and Lim, \{Kian Ming\} and \{Yan Lim\}, Jit",

note = "Publisher Copyright: {\textcopyright} 2023 IEEE.; International Conference on Information and Communication Technology 2023, ICoICT 2023 ; Conference date: 23-08-2023 Through 24-08-2023",

year = "2023",

doi = "10.1109/ICoICT58202.2023.10262721",

language = "English",

isbn = "9798350333039",

pages = "137--141",

editor = "Lee-Ying Chong and Tee Connie and \{Dwi Jatmiko Suwawi\}, Dawam and \{Liang Tan\}, Joon",

booktitle = "11th International Conference on Information and Communication Technology, ICoICT 2023",

publisher = "IEEE, Institute of Electrical and Electronics Engineers",

address = "United States of America",

url = "https://ieeexplore.ieee.org/xpl/conhome/10262402/proceeding, https://www.google.com/url?sa=t\&rct=j\&q=\&esrc=s\&source=web\&cd=\&ved=2ahUKEwiKrtbNp-GEAxUbs1YBHZlDAVkQFnoECA8QAQ\&url=https\%3A\%2F\%2Fwww.icoict.org\%2F2023-icoict\%2F\&usg=AOvVaw2TRUvwZYzGEHUrk65UNNeI\&opi=89978449",

}

Mook, CE, Poo Lee, C, Lim, KM & Yan Lim, J 2023, Handwritten Character and Digit Recognition with Deep Convolutional Neural Networks: A Comparative Study. in L-Y Chong, T Connie, D Dwi Jatmiko Suwawi & J Liang Tan (eds), 11th International Conference on Information and Communication Technology, ICoICT 2023. IEEE, Institute of Electrical and Electronics Engineers, Piscataway NJ USA, pp. 137-141, International Conference on Information and Communication Technology 2023, Melaka, Malaysia, 23/08/23. https://doi.org/10.1109/ICoICT58202.2023.10262721

Handwritten Character and Digit Recognition with Deep Convolutional Neural Networks: A Comparative Study. / Mook, Chui En; Poo Lee, Chin; Lim, Kian Ming et al.
11th International Conference on Information and Communication Technology, ICoICT 2023. ed. / Lee-Ying Chong; Tee Connie; Dawam Dwi Jatmiko Suwawi; Joon Liang Tan. Piscataway NJ USA: IEEE, Institute of Electrical and Electronics Engineers, 2023. p. 137-141.

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research

TY - GEN

T1 - Handwritten Character and Digit Recognition with Deep Convolutional Neural Networks

T2 - International Conference on Information and Communication Technology 2023

AU - Mook, Chui En

AU - Poo Lee, Chin

AU - Lim, Kian Ming

AU - Yan Lim, Jit

N1 - Conference code: 11th

PY - 2023

Y1 - 2023

N2 - Handwritten character or digit recognition involves automatically classifying handwritten characters or digits from images. Previous studies focused on specific datasets and did not thoroughly compare different CNN architectures. This paper addresses these limitations by presenting a comparative study of six popular CNN architectures (VGG16, Xception, ResNet152V2, InceptionResNetV2, MobileNetV2, and DenseNet201) on three diverse datasets: English Handwritten Characters, Handwritten Digits, and MNIST. The experimental results demonstrate that the InceptionResNetV2 model with data augmentation achieves the highest accuracy across all datasets, with accuracies of 93.26%, 97.16%, and 99.71% on the English Handwritten Characters, Handwritten Digits, and MNIST datasets, respectively.

AB - Handwritten character or digit recognition involves automatically classifying handwritten characters or digits from images. Previous studies focused on specific datasets and did not thoroughly compare different CNN architectures. This paper addresses these limitations by presenting a comparative study of six popular CNN architectures (VGG16, Xception, ResNet152V2, InceptionResNetV2, MobileNetV2, and DenseNet201) on three diverse datasets: English Handwritten Characters, Handwritten Digits, and MNIST. The experimental results demonstrate that the InceptionResNetV2 model with data augmentation achieves the highest accuracy across all datasets, with accuracies of 93.26%, 97.16%, and 99.71% on the English Handwritten Characters, Handwritten Digits, and MNIST datasets, respectively.

KW - Convolution Neural Network

KW - Data Augmentation

KW - Handwritten Character Recognition

KW - Handwritten Digit Recognition

UR - https://www.scopus.com/pages/publications/85174394586

U2 - 10.1109/ICoICT58202.2023.10262721

DO - 10.1109/ICoICT58202.2023.10262721

M3 - Conference Paper

AN - SCOPUS:85174394586

SN - 9798350333039

SP - 137

EP - 141

BT - 11th International Conference on Information and Communication Technology, ICoICT 2023

A2 - Chong, Lee-Ying

A2 - Connie, Tee

A2 - Dwi Jatmiko Suwawi, Dawam

A2 - Liang Tan, Joon

PB - IEEE, Institute of Electrical and Electronics Engineers

CY - Piscataway NJ USA

Y2 - 23 August 2023 through 24 August 2023

ER -

Mook CE, Poo Lee C, Lim KM, Yan Lim J. Handwritten Character and Digit Recognition with Deep Convolutional Neural Networks: A Comparative Study. In Chong LY, Connie T, Dwi Jatmiko Suwawi D, Liang Tan J, editors, 11th International Conference on Information and Communication Technology, ICoICT 2023. Piscataway NJ USA: IEEE, Institute of Electrical and Electronics Engineers. 2023. p. 137-141 doi: 10.1109/ICoICT58202.2023.10262721

Handwritten Character and Digit Recognition with Deep Convolutional Neural Networks: A Comparative Study

Abstract

Conference

Keywords

Access to Document

Other files and links

Cite this