Handwritten Bangla numeral recognition using deep long short term memory

Recognition of handwritten numerals has gained much interest in recent years due to its various application potentials. Bangla is a major language in Indian subcontinent and is the first language of Bangladesh; but unfortunately, study regarding handwritten Bangla numeral recognition (HBNR) is very...

Full description

Bibliographic Details
Main Authors: Ahmed, Mahtab, Akhand, M. A. H, Rahman, M.M. Hafizur
Format: Proceeding Paper
Language:English
English
Published: Institute of Electrical and Electronics Engineers Inc. ( IEEE) 2017
Subjects:
Online Access:http://irep.iium.edu.my/54935/
http://irep.iium.edu.my/54935/1/54935_Handwritten%20Bangla%20numeral%20recognition.pdf
http://irep.iium.edu.my/54935/2/54935_Handwritten%20Bangla%20numeral%20recognition_SCOPUS.pdf
_version_ 1848784493748420608
author Ahmed, Mahtab
Akhand, M. A. H
Rahman, M.M. Hafizur
author_facet Ahmed, Mahtab
Akhand, M. A. H
Rahman, M.M. Hafizur
author_sort Ahmed, Mahtab
building IIUM Repository
collection Online Access
description Recognition of handwritten numerals has gained much interest in recent years due to its various application potentials. Bangla is a major language in Indian subcontinent and is the first language of Bangladesh; but unfortunately, study regarding handwritten Bangla numeral recognition (HBNR) is very few with respect to other major languages such as English, Roman etc. Some noteworthy research works have been conducted for recognition of Bangla handwritten numeral using artificial neural network (ANN) as ANN and its various updated models are found efficient for classification task. The aim of this study is to develop a better Bangla handwritten numeral recognition system and hence investigated deep architecture of Long Short Term Memory (LSTM) method. LSTM is a variant of recurrent neural networks (RNN) and is applied efficiently for image classification with its distinct features. The proposed HBNR-LSTM normalizes the written numeral images first and then employs two layers of LSTM to classify individual numerals. Unlike other methods, it does not employ any feature extraction technique. Benchmark dataset with 22000 hand written numerals with different shapes, sizes and variations are used in this study. The proposed method is shown satisfactory recognition accuracy and outperformed other prominent exiting methods.
first_indexed 2025-11-14T16:38:08Z
format Proceeding Paper
id iium-54935
institution International Islamic University Malaysia
institution_category Local University
language English
English
last_indexed 2025-11-14T16:38:08Z
publishDate 2017
publisher Institute of Electrical and Electronics Engineers Inc. ( IEEE)
recordtype eprints
repository_type Digital Repository
spelling iium-549352017-06-05T01:01:52Z http://irep.iium.edu.my/54935/ Handwritten Bangla numeral recognition using deep long short term memory Ahmed, Mahtab Akhand, M. A. H Rahman, M.M. Hafizur TK7800 Electronics. Computer engineering. Computer hardware. Photoelectronic devices Recognition of handwritten numerals has gained much interest in recent years due to its various application potentials. Bangla is a major language in Indian subcontinent and is the first language of Bangladesh; but unfortunately, study regarding handwritten Bangla numeral recognition (HBNR) is very few with respect to other major languages such as English, Roman etc. Some noteworthy research works have been conducted for recognition of Bangla handwritten numeral using artificial neural network (ANN) as ANN and its various updated models are found efficient for classification task. The aim of this study is to develop a better Bangla handwritten numeral recognition system and hence investigated deep architecture of Long Short Term Memory (LSTM) method. LSTM is a variant of recurrent neural networks (RNN) and is applied efficiently for image classification with its distinct features. The proposed HBNR-LSTM normalizes the written numeral images first and then employs two layers of LSTM to classify individual numerals. Unlike other methods, it does not employ any feature extraction technique. Benchmark dataset with 22000 hand written numerals with different shapes, sizes and variations are used in this study. The proposed method is shown satisfactory recognition accuracy and outperformed other prominent exiting methods. Institute of Electrical and Electronics Engineers Inc. ( IEEE) 2017-01-16 Proceeding Paper PeerReviewed application/pdf en http://irep.iium.edu.my/54935/1/54935_Handwritten%20Bangla%20numeral%20recognition.pdf application/pdf en http://irep.iium.edu.my/54935/2/54935_Handwritten%20Bangla%20numeral%20recognition_SCOPUS.pdf Ahmed, Mahtab and Akhand, M. A. H and Rahman, M.M. Hafizur (2017) Handwritten Bangla numeral recognition using deep long short term memory. In: 2016 6th International Conference on Information and Communication Technology for The Muslim World, Nov. 22 ~ 24, 2016, Jakarta, Indonesia. http://ieeexplore.ieee.org/document/7814922/ 10.1109/ICT4M.2016.62
spellingShingle TK7800 Electronics. Computer engineering. Computer hardware. Photoelectronic devices
Ahmed, Mahtab
Akhand, M. A. H
Rahman, M.M. Hafizur
Handwritten Bangla numeral recognition using deep long short term memory
title Handwritten Bangla numeral recognition using deep long short term memory
title_full Handwritten Bangla numeral recognition using deep long short term memory
title_fullStr Handwritten Bangla numeral recognition using deep long short term memory
title_full_unstemmed Handwritten Bangla numeral recognition using deep long short term memory
title_short Handwritten Bangla numeral recognition using deep long short term memory
title_sort handwritten bangla numeral recognition using deep long short term memory
topic TK7800 Electronics. Computer engineering. Computer hardware. Photoelectronic devices
url http://irep.iium.edu.my/54935/
http://irep.iium.edu.my/54935/
http://irep.iium.edu.my/54935/
http://irep.iium.edu.my/54935/1/54935_Handwritten%20Bangla%20numeral%20recognition.pdf
http://irep.iium.edu.my/54935/2/54935_Handwritten%20Bangla%20numeral%20recognition_SCOPUS.pdf