International Journal of Engineering Research in Electronics and Communication Engineering

Handwritten Devanagari Compound Character Recognition Using Zernike Moment

Author : Juhee Sachdeva ¹ Dr. Shilpa Kulkarni ²

Date of Publication :20th December 2017

Abstract: Character recognition of Devanagari compound characters is a difficult task because characters are written in half forms of original characters. In this paper, we present OCR for Handwritten Devanagari compound character. Recognition of Devanagari compound characters consists of three main phases Preprocessing, Feature Extraction and Classification. During last few decades many research has been conducted to find a solution for Character recognition as it is a solution for many real world problems. The major problem for processing large volume of data can be solved. Work has been done in recognizing handwritten characters in Languages like English,Chinese,Japanese and Arabic. This paper focuses on the recognition of Devanagari compound characters. The feature extraction stage uses Zernike Moments based feature extraction technique that analyzes and select set of features that can be used to uniquely identify the character. Zernike Moment based techniques uses rotation invariance that is successfully applied to many Pattern Recognition problems and character recognition system. Moment based system uses global features that is suitable for compound character recognition. For classification phase, Support Vector Machine is used . The system reported a high recognition rate of accuracy.

Reference :

1. Optical Character Recognition by Rice, Stephen V., Nagy, George, Nartker, Thomas A Springer Science & Business Media, 31-May-1999.
2. Gonzalez, Rafael C., and Richard E. Woods. "Digital image processing." (2002).
3. Cheriet, Mohamed, et al. Character recognition systems: a guide for students and practitioners. John Wiley & Sons, 2007.
4. H.Bunke P.S.P Wang-"Handbook of character recognition and document image analysis." (1997).
5. Connell, Scott D., R. M. K. Sinha, and Anil K. Jain. "Recognition of unconstrained online Devanagari characters." Pattern Recognition, 2000. Proceedings. 15th International Conference on. Vol. 2. IEEE, 2000.
6. Bansal, Veena, and R. M. K. Sinha. "On how to describe shapes of Devanagari characters and use them for recognition." Document Analysis and Recognition, 1999. ICDAR'99. Proceedings of the Fifth International Conference on. IEEE, 1999.