A Mixture of MLPNN/HMM to Demonstrate the Procedure for Online Hindi Writing Recognition-Reference-Cited by-同舟云学术

A Mixture of MLPNN/HMM to Demonstrate the Procedure for Online Hindi Writing Recognition

Published:2022-05-27 Issue: Volume: Page:414-425
ISSN:2581-6942
Container-title:International Journal of Case Studies in Business, IT, and Education
language:en
Short-container-title:IJCSBE

Author:

Patil Vinita¹,Aithal P. S.²

Affiliation:

1. Post- Doctoral Research Scholar, Srinivas University, Mangalore, India.

2. Professor, College of Management & Commerce, Srinivas University, Mangalore, India.

Abstract

Purpose: This presented work demonstrates a mixture of the MLPNN/HMM model for operational Hindi writing recognition. The planned structure is proposed on behalf of HMMs and various-Layer Perceptron Neural Linkages. Signal of input is generated toward uninterrupted knocks named segments constructed proceeding of the Elliptical approach through scrutinizing the extremisms arguments of the rounded rapidity outline. The linkage of neural accomplished through subdivision near relative evidence stays castoff near-abstract course atmosphere prospects. Proposed outcomes of that system remain translated through HMMs towards run after attractiveness of equal credit. Now appraisals are scheduled to DBMs Hindi database which proficient 97.8% attractiveness appreciation exactness that signifies statistically significantly imperative trendy evaluation through attractiveness appreciation exactitudes attained after advanced online Hindi systems. Design/Methodology/Approach: Developing a mixture of the MLPNN/HMM model for operational Hindi writing recognition. Findings/Result: Based on the proposed model that system remains translated through HMMs towards run after attractiveness of equal credit. Now appraisals are scheduled to DBMs Hindi database which proficient 97.8% attractiveness appreciation exactness that signifies statistically significantly imperative trendy evaluation through attractiveness appreciation exactitudes attained after advanced online Hindi systems. Originality/Value: A new conceptual mixture of the MLPNN/HMM model has been proposed for operations of Hindi writing recognition techniques based on Perceptron Neural Linkages. Paper Type: Conceptual Research.

Publisher

Srinivas University

Subject

General Medicine

Reference26 articles.

1. Günter, S., & Bunke, H. (2004). HMM-based handwritten word recognition: on the optimization of the number of states, training iterations and Gaussian components. Pattern Recognition, 37(1), 2069–2079. https://dl.acm.org/doi/abs/10.1016/j.patcog.2004.04.006

2. Alimi, A. M. (1997). An Evolutionary neuro-fuzzy approach to recognize on –line Arabic handwriting. Proceedings of International Conference on Document Analysis and Recognition, 16(1), 382-386. DOI:10.1109/ICDAR.1997.619875

3. Graves, A., Fernández, S., Gomez, F., & Schmidhube, J. (2006) Connectionist temporal classification: labeling unsegmented sequence data with recurrent neural networks. In Proc. Int. Conf. on Machine Learning, pp. 369–376. https://dl.acm.org/doi/10.1145/1143844.1143891

4. Graves, A., Liwicki, M. Fernández, S., Bertolami, R., Bunke, H., & Schmidhuber, J. (2009). A novel connectionist system for unconstrained handwriting recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence, 31(5), 855–868. DOI: 10.1109/TPAMI.2008.137

5. Biadsy, F., El-Sana, J., & Habash, N. (2006). Online Arabic Handwriting Recognition Using Hidden Markov Models. Proc. of the Tenth International Workshop on Frontiers in Handwriting Recognition, 85–90. https://hal.archives-ouvertes.fr/inria-00108306/