Digital Image Processing

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

Classification and Segmentation of Telugu Touching Characters


Affiliations
1 Department of Electronics and Communication Engineering, Deccan College of Engineering and Technology, Hyderabad, India
2 Department of Electronics and Communication nEngineering, JNTU College of Engineering, Hyderabad, India
     

   Subscribe/Renew Journal


Improper binarization, inadequacies in printing, orthography of the script, font types, and low scanning resolution are some of the factors which contribute to the touching characters in OCR system. In this paper touching characters of Telugu script are classified based on the location of touching and two algorithms for segmentation of touching characters in top zone and middle zone are proposed. This makes it easier to segment them with appropriate strategies for each class. Characteristics of projection profiles and side profiles are used to identify and segment the touching characters. The success rates of 92.67% and 91.28% are achieved for segmenting the touching characters in top zone and middle zone respectively.

Keywords

Classification of Touching Characters, Side Profiles, Touching Characters in Middle Zone, Touching Characters in Top Zone.
User
Subscription Login to verify subscription
Notifications
Font Size

Abstract Views: 221

PDF Views: 3




  • Classification and Segmentation of Telugu Touching Characters

Abstract Views: 221  |  PDF Views: 3

Authors

J. Bharathi
Department of Electronics and Communication Engineering, Deccan College of Engineering and Technology, Hyderabad, India
P. Chandrasekhar Reddy
Department of Electronics and Communication nEngineering, JNTU College of Engineering, Hyderabad, India

Abstract


Improper binarization, inadequacies in printing, orthography of the script, font types, and low scanning resolution are some of the factors which contribute to the touching characters in OCR system. In this paper touching characters of Telugu script are classified based on the location of touching and two algorithms for segmentation of touching characters in top zone and middle zone are proposed. This makes it easier to segment them with appropriate strategies for each class. Characteristics of projection profiles and side profiles are used to identify and segment the touching characters. The success rates of 92.67% and 91.28% are achieved for segmenting the touching characters in top zone and middle zone respectively.

Keywords


Classification of Touching Characters, Side Profiles, Touching Characters in Middle Zone, Touching Characters in Top Zone.