A segmentation based optical character recognition system for Bangla printed text.

Saved in:

Bibliographic Details
Title:	A segmentation based optical character recognition system for Bangla printed text.
Authors:	Mahbub, Mahir^1,2 mahbub0001@bdu.ac.bd, Kabir, Ahmedul² kabir@iit.du.ac.bd
Source:	Telkomnika. Jun2026, Vol. 24 Issue 3, p945-956. 12p.
Subjects:	Optical character recognition, Convolutional neural networks, Image segmentation, Pattern recognition systems, Computational linguistics
Geographic Terms:	Bangladesh
Abstract:	Bangla ranks as the fifth most spoken language globally, catalyzing significant interest in the development of Bangla optical character recognition (OCR) systems. The intricate structure of the Bangla script, including compound characters, modifiers, and headlines, complicates the formation of words. This research introduces a complete OCR system pipeline for printed Bangla text. It employs a thinning-based segmentation approach combined with a convolutional neural network (CNN) to recognize Bangla fonts. Additionally, a part of speech (POS)-aware spell checker is proposed that automatically corrects misspelled words while considering their context within the sentence. We introduce semi-generalized filters that adapt to new fonts, addressing conjunct formation challenges in Bangla OCR. This flexible design allows for adaptation to new fonts. The ResNet50 model is utilized to accurately recognize segmented characters and modifiers. We achieve a character segmentation error of 3.354% and an overall segmentation error of 2.332%. The ResNet50 recognition model achieves an accuracy of 98.345%. [ABSTRACT FROM AUTHOR]
	Copyright of Telkomnika is the property of Department of Electrical Engineering, Ahmad Dahlan University and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database:	Engineering Source

Description
Abstract:	Bangla ranks as the fifth most spoken language globally, catalyzing significant interest in the development of Bangla optical character recognition (OCR) systems. The intricate structure of the Bangla script, including compound characters, modifiers, and headlines, complicates the formation of words. This research introduces a complete OCR system pipeline for printed Bangla text. It employs a thinning-based segmentation approach combined with a convolutional neural network (CNN) to recognize Bangla fonts. Additionally, a part of speech (POS)-aware spell checker is proposed that automatically corrects misspelled words while considering their context within the sentence. We introduce semi-generalized filters that adapt to new fonts, addressing conjunct formation challenges in Bangla OCR. This flexible design allows for adaptation to new fonts. The ResNet50 model is utilized to accurately recognize segmented characters and modifiers. We achieve a character segmentation error of 3.354% and an overall segmentation error of 2.332%. The ResNet50 recognition model achieves an accuracy of 98.345%. [ABSTRACT FROM AUTHOR]
ISSN:	16936930
DOI:	10.12928/TELKOMNIKA.v24i3.26961