Open Access Open Access  Restricted Access Subscription Access
Open Access Open Access Open Access  Restricted Access Restricted Access Subscription Access

Separation of Tamil and Devanagari Script Words in Printed Bilingual Document Images


Affiliations
1 Department of Computer Science, Anna University, Chennai, India
2 Department of Computer Science, College of Engineering, Anna University, Chennai, India
3 Computer Science and Engineering Department, College of Engineering, Anna University, Guindy, India
     

   Subscribe/Renew Journal


Identification of scripts from bi-script document is one of the important steps in the design of an OCR system for successful analysis and recognition. Most optical character recognition (OCR) systems can recognize at most a few scripts. But for large archives of document images that contain different scripts, there must be some way to automatically categorize these documents before applying the proper OCR on them. Much work has already been reported in this area. In the Indian context, though some results have been reported, the task is still at its infancy. This paper presents a research in the identification of Tamil, Devanagari scripts at word level irrespective of their font faces and sizes. The proposed technique performs document vectorization method which generates vectors from the nine zones segmented over the characters based on their shape, density and transition features. Then script is proposed technique identifies scripts with minimal pre-processing and high accuracy. It can also be extended for other scripts. Since this determined by using Rule based classifiers containing set of classification rules which are raised from the vectors. Results from experiments, simulations, and human vision encounter that the system can act as a plug-in, this can be embedded with OCR prior to the recognition stage.

Keywords

Bi-Lingual Document, Script Identification, Rule Based Classification, Optical Character Recognition (OCR).
User
Subscription Login to verify subscription
Notifications
Font Size

Abstract Views: 241

PDF Views: 4




  • Separation of Tamil and Devanagari Script Words in Printed Bilingual Document Images

Abstract Views: 241  |  PDF Views: 4

Authors

R. Rathinapriya
Department of Computer Science, Anna University, Chennai, India
S. Abirami
Department of Computer Science, College of Engineering, Anna University, Chennai, India
B. Manjula
Computer Science and Engineering Department, College of Engineering, Anna University, Guindy, India

Abstract


Identification of scripts from bi-script document is one of the important steps in the design of an OCR system for successful analysis and recognition. Most optical character recognition (OCR) systems can recognize at most a few scripts. But for large archives of document images that contain different scripts, there must be some way to automatically categorize these documents before applying the proper OCR on them. Much work has already been reported in this area. In the Indian context, though some results have been reported, the task is still at its infancy. This paper presents a research in the identification of Tamil, Devanagari scripts at word level irrespective of their font faces and sizes. The proposed technique performs document vectorization method which generates vectors from the nine zones segmented over the characters based on their shape, density and transition features. Then script is proposed technique identifies scripts with minimal pre-processing and high accuracy. It can also be extended for other scripts. Since this determined by using Rule based classifiers containing set of classification rules which are raised from the vectors. Results from experiments, simulations, and human vision encounter that the system can act as a plug-in, this can be embedded with OCR prior to the recognition stage.

Keywords


Bi-Lingual Document, Script Identification, Rule Based Classification, Optical Character Recognition (OCR).



DOI: https://doi.org/10.36039/ciitaas%2F1%2F1%2F2009%2F107050.15-21