Abstract
In recognizing the handwritten Sanskrit characters, several procedures have to be performed; no single machine or single process can perform in Sanskrit character recognition. The issues of handwritten character recognition are still an active area of research, with ever-improving the need for automation of office work, it is vital to give effective and practical solutions. Initially, the scanned documents are preprocessed using the balanced histogram process for binarization and non-local median filter for denoizing process. Next multiscale segmentation process is carried out to extract the features. A comparison of the character recognition classifiers Naive Bayesian classifiers, SVM classifier and CNN classifier is done to analyze the efficiency of the classifiers. The accuracy of the CNN classifier is 93.45% to recognize the given Sanskrit text.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Agarwal, P.: Hand-written character recognition using Kohonen network. IJCST 2(3), 10–18 (2011)
Alirezaee, S., Aghaeinia, H., Faez, K., Fard, A.S.: An efficient feature extraction method for the middle-age character recognition. Lect. Notes Comput. Sci. 3645(PART II), 998–1006 (2005)
Ameri, M.R., Haji, M., Fischer, A., Ponson, D., Bui, T.D.: A feature extraction method for cursive character recognition using higher-order singular value decomposition. In: 2014 14th International Conference on Frontiers in Handwriting Recognition, pp. 512–516 (2014)
Dwivedi, N., Srivastava, K., Arya, N.: Sanskrit Word Recognition Using Prewitt’s Operator and Support Vector Classification. ICECCN, pp. 265–269 (2013)
El Qacimy, B., Kerroum, M.A., Hammouch, A.: Feature extraction based on DCT for handwritten digit recognition. Int. J. Comput. Sci. Issues (IJCSI) 11(6), 27–27 (2014)
Joshi, M.R., Sabale, M.V.V.: Offline Character Recognition for Printed Text in Devanagari Using Neural Network and Genetic Algorithm, pp. 1–8 (2013)
Kale, K.V., Deshmukh, P.D., Chavan, S.V., Kazi, M.M., Rode, Y.S.: Zernike moment feature extraction for handwritten Devanagari compound character recognition. Sci. Inf. Conf. 3(1), 459–466 (2013)
Liu, X.Y., Blumenstein, M.: Experimental analysis of the modified direction feature for cursive character recognition. In: Proceedings—International Workshop on Frontiers in Handwriting Recognition, IWFHR, pp. 353–358 (2004)
Dineshkumar, R., Suganthi, J.: Sanskrit character recognition system using neural network. Indian J. Sci. Technol. 8(1), 65–69. E-ISSN: 0974-5645 (2015)
Dineshkumar, R.: Offline Sanskrit handwritten character recognition framework based on multilayer feed forward network with intelligent character recognition. Asian J. Inf. Technol. 15(11), 1678–1685. ISSN: 1682-3915 (2016)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Dinesh Kumar, R., Kalimuthu, M., Jayaram, B. (2022). Character Recognition System Using CNN for Sanskrit Text. In: Satyanarayana, C., Gao, XZ., Ting, CY., Muppalaneni, N.B. (eds) Proceedings of the International Conference on Computer Vision, High Performance Computing, Smart Devices and Networks. Advanced Technologies and Societal Change. Springer, Singapore. https://doi.org/10.1007/978-981-19-4044-6_1
Download citation
DOI: https://doi.org/10.1007/978-981-19-4044-6_1
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-19-4043-9
Online ISBN: 978-981-19-4044-6
eBook Packages: Computer ScienceComputer Science (R0)