A New Model Evaluation Framework for Tamil Handwritten Character Recognition

B. R. Kavitha; Noushath Shaffi; Mufti Mahmud; Faizal Hajamohideen; Priyalakshmi Narayanan

doi:10.1007/978-981-99-1916-1_7

A New Model Evaluation Framework for Tamil Handwritten Character Recognition

B. R. Kavitha^*
, Noushath Shaffi
, Mufti Mahmud
, Faizal Hajamohideen
, Priyalakshmi Narayanan

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

The robustness of any pattern recognition model relies heavily on the availability of comprehensive samples. Until last year, the Tamil Handwritten Character Recognition (HWCR) works relied on the solitary HPL Tamil dataset [1]. Recently, a new benchmarking for Tamil HWCR was published [2] comprising 94000 samples in total. The efficiency of corroboration using multiple standardized databases is crucial in advancing any research area. Towards this aim, in this paper, we showed different ways of experimentation with these datasets. For this purpose, we utilized transfer learning, and a custom deep neural network, a recently published work for Tamil HWCR [3]. Different experimental setups were suggested that involved independent, cross-testing, and mixed modes of model building and evaluation using two standardized datasets. These setups form a rigorous testing framework for analyzing Tamil HWCR tasks. The work presented in this paper is the first to report the results of Tamil HWCR using two standardized datasets and sets a new model evaluation benchmark. For rapid reproducibility and dissemination, the code and materials used in this study are available at https://github.com/Kavitha-BR-VIT/Tamil-HWCR.

Original language	English
Title of host publication	Proceedings of Trends in Electronics and Health Informatics - TEHI 2022
Editors	Mufti Mahmud, Claudia Mendoza-Barrera, M. Shamim Kaiser, Anirban Bandyopadhyay, Kanad Ray, Eduardo Lugo
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	93-106
Number of pages	14
ISBN (Print)	9789819919154
DOIs	https://doi.org/10.1007/978-981-99-1916-1_7
State	Published - 2023
Externally published	Yes
Event	2nd International Conference on Trends in Electronics and Health Informatics, TEHI 2022 - Puebla, Mexico Duration: 7 Dec 2022 → 9 Dec 2022

Publication series

Name	Lecture Notes in Networks and Systems
Volume	675 LNNS
ISSN (Print)	2367-3370
ISSN (Electronic)	2367-3389

Conference

Conference	2nd International Conference on Trends in Electronics and Health Informatics, TEHI 2022
Country/Territory	Mexico
City	Puebla
Period	7/12/22 → 9/12/22

Bibliographical note

Publisher Copyright:
© 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Keywords

Convolution neural network
Explainable OCR
Handwritten tamil datasets
OCR
Pretrained models
Tamil handwritten character recognition

ASJC Scopus subject areas

Control and Systems Engineering
Signal Processing
Computer Networks and Communications

Access to Document

10.1007/978-981-99-1916-1_7

Cite this

Kavitha, B. R., Shaffi, N., Mahmud, M., Hajamohideen, F., & Narayanan, P. (2023). A New Model Evaluation Framework for Tamil Handwritten Character Recognition. In M. Mahmud, C. Mendoza-Barrera, M. S. Kaiser, A. Bandyopadhyay, K. Ray, & E. Lugo (Eds.), Proceedings of Trends in Electronics and Health Informatics - TEHI 2022 (pp. 93-106). (Lecture Notes in Networks and Systems; Vol. 675 LNNS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-981-99-1916-1_7

Kavitha, B. R. ; Shaffi, Noushath ; Mahmud, Mufti et al. / A New Model Evaluation Framework for Tamil Handwritten Character Recognition. Proceedings of Trends in Electronics and Health Informatics - TEHI 2022. editor / Mufti Mahmud ; Claudia Mendoza-Barrera ; M. Shamim Kaiser ; Anirban Bandyopadhyay ; Kanad Ray ; Eduardo Lugo. Springer Science and Business Media Deutschland GmbH, 2023. pp. 93-106 (Lecture Notes in Networks and Systems).

@inproceedings{0a3b79bef73249309a9566cde5e6c174,

title = "A New Model Evaluation Framework for Tamil Handwritten Character Recognition",

abstract = "The robustness of any pattern recognition model relies heavily on the availability of comprehensive samples. Until last year, the Tamil Handwritten Character Recognition (HWCR) works relied on the solitary HPL Tamil dataset [1]. Recently, a new benchmarking for Tamil HWCR was published [2] comprising 94000 samples in total. The efficiency of corroboration using multiple standardized databases is crucial in advancing any research area. Towards this aim, in this paper, we showed different ways of experimentation with these datasets. For this purpose, we utilized transfer learning, and a custom deep neural network, a recently published work for Tamil HWCR [3]. Different experimental setups were suggested that involved independent, cross-testing, and mixed modes of model building and evaluation using two standardized datasets. These setups form a rigorous testing framework for analyzing Tamil HWCR tasks. The work presented in this paper is the first to report the results of Tamil HWCR using two standardized datasets and sets a new model evaluation benchmark. For rapid reproducibility and dissemination, the code and materials used in this study are available at https://github.com/Kavitha-BR-VIT/Tamil-HWCR.",

keywords = "Convolution neural network, Explainable OCR, Handwritten tamil datasets, OCR, Pretrained models, Tamil handwritten character recognition",

author = "Kavitha, \{B. R.\} and Noushath Shaffi and Mufti Mahmud and Faizal Hajamohideen and Priyalakshmi Narayanan",

note = "Publisher Copyright: {\textcopyright} 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.; 2nd International Conference on Trends in Electronics and Health Informatics, TEHI 2022 ; Conference date: 07-12-2022 Through 09-12-2022",

year = "2023",

doi = "10.1007/978-981-99-1916-1\_7",

language = "English",

isbn = "9789819919154",

series = "Lecture Notes in Networks and Systems",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "93--106",

editor = "Mufti Mahmud and Claudia Mendoza-Barrera and Kaiser, \{M. Shamim\} and Anirban Bandyopadhyay and Kanad Ray and Eduardo Lugo",

booktitle = "Proceedings of Trends in Electronics and Health Informatics - TEHI 2022",

address = "Germany",

}

Kavitha, BR, Shaffi, N, Mahmud, M, Hajamohideen, F & Narayanan, P 2023, A New Model Evaluation Framework for Tamil Handwritten Character Recognition. in M Mahmud, C Mendoza-Barrera, MS Kaiser, A Bandyopadhyay, K Ray & E Lugo (eds), Proceedings of Trends in Electronics and Health Informatics - TEHI 2022. Lecture Notes in Networks and Systems, vol. 675 LNNS, Springer Science and Business Media Deutschland GmbH, pp. 93-106, 2nd International Conference on Trends in Electronics and Health Informatics, TEHI 2022, Puebla, Mexico, 7/12/22. https://doi.org/10.1007/978-981-99-1916-1_7

A New Model Evaluation Framework for Tamil Handwritten Character Recognition. / Kavitha, B. R.; Shaffi, Noushath; Mahmud, Mufti et al.
Proceedings of Trends in Electronics and Health Informatics - TEHI 2022. ed. / Mufti Mahmud; Claudia Mendoza-Barrera; M. Shamim Kaiser; Anirban Bandyopadhyay; Kanad Ray; Eduardo Lugo. Springer Science and Business Media Deutschland GmbH, 2023. p. 93-106 (Lecture Notes in Networks and Systems; Vol. 675 LNNS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - A New Model Evaluation Framework for Tamil Handwritten Character Recognition

AU - Kavitha, B. R.

AU - Shaffi, Noushath

AU - Mahmud, Mufti

AU - Hajamohideen, Faizal

AU - Narayanan, Priyalakshmi

PY - 2023

Y1 - 2023

N2 - The robustness of any pattern recognition model relies heavily on the availability of comprehensive samples. Until last year, the Tamil Handwritten Character Recognition (HWCR) works relied on the solitary HPL Tamil dataset [1]. Recently, a new benchmarking for Tamil HWCR was published [2] comprising 94000 samples in total. The efficiency of corroboration using multiple standardized databases is crucial in advancing any research area. Towards this aim, in this paper, we showed different ways of experimentation with these datasets. For this purpose, we utilized transfer learning, and a custom deep neural network, a recently published work for Tamil HWCR [3]. Different experimental setups were suggested that involved independent, cross-testing, and mixed modes of model building and evaluation using two standardized datasets. These setups form a rigorous testing framework for analyzing Tamil HWCR tasks. The work presented in this paper is the first to report the results of Tamil HWCR using two standardized datasets and sets a new model evaluation benchmark. For rapid reproducibility and dissemination, the code and materials used in this study are available at https://github.com/Kavitha-BR-VIT/Tamil-HWCR.

AB - The robustness of any pattern recognition model relies heavily on the availability of comprehensive samples. Until last year, the Tamil Handwritten Character Recognition (HWCR) works relied on the solitary HPL Tamil dataset [1]. Recently, a new benchmarking for Tamil HWCR was published [2] comprising 94000 samples in total. The efficiency of corroboration using multiple standardized databases is crucial in advancing any research area. Towards this aim, in this paper, we showed different ways of experimentation with these datasets. For this purpose, we utilized transfer learning, and a custom deep neural network, a recently published work for Tamil HWCR [3]. Different experimental setups were suggested that involved independent, cross-testing, and mixed modes of model building and evaluation using two standardized datasets. These setups form a rigorous testing framework for analyzing Tamil HWCR tasks. The work presented in this paper is the first to report the results of Tamil HWCR using two standardized datasets and sets a new model evaluation benchmark. For rapid reproducibility and dissemination, the code and materials used in this study are available at https://github.com/Kavitha-BR-VIT/Tamil-HWCR.

KW - Convolution neural network

KW - Explainable OCR

KW - Handwritten tamil datasets

KW - OCR

KW - Pretrained models

KW - Tamil handwritten character recognition

UR - https://www.scopus.com/pages/publications/85164931836

U2 - 10.1007/978-981-99-1916-1_7

DO - 10.1007/978-981-99-1916-1_7

M3 - Conference contribution

AN - SCOPUS:85164931836

SN - 9789819919154

T3 - Lecture Notes in Networks and Systems

SP - 93

EP - 106

BT - Proceedings of Trends in Electronics and Health Informatics - TEHI 2022

A2 - Mahmud, Mufti

A2 - Mendoza-Barrera, Claudia

A2 - Kaiser, M. Shamim

A2 - Bandyopadhyay, Anirban

A2 - Ray, Kanad

A2 - Lugo, Eduardo

PB - Springer Science and Business Media Deutschland GmbH

T2 - 2nd International Conference on Trends in Electronics and Health Informatics, TEHI 2022

Y2 - 7 December 2022 through 9 December 2022

ER -

Kavitha BR, Shaffi N, Mahmud M, Hajamohideen F, Narayanan P. A New Model Evaluation Framework for Tamil Handwritten Character Recognition. In Mahmud M, Mendoza-Barrera C, Kaiser MS, Bandyopadhyay A, Ray K, Lugo E, editors, Proceedings of Trends in Electronics and Health Informatics - TEHI 2022. Springer Science and Business Media Deutschland GmbH. 2023. p. 93-106. (Lecture Notes in Networks and Systems). doi: 10.1007/978-981-99-1916-1_7