A Study on Handwritten Character Recognition Mechanisms using Neural Networks

  • Sunitha S Nair, Dr. P. Ranjit Jeba Thangaiah

Abstract

Handwritten character recognition grabs the attention of several researchers in recent times. Character Recognition is one of the recent tasks that relates to different forms of handwritten text based image and digits data analysis. The analysis of recognizing the similar patterns for the application requirements encouraged by the feature analytic schemes and classification models.  Neural networks algorithms are widely adopted by the researchers for its attentive and recurrent training network modules. Despite application requirements, the extraction levels of features, so as to train the neural network classifiers depends on several constraints. The usage of high cardinal features will increase the computational complexity. Therefore, it is suggested that an optimized cardinal of features will not affect the boundaries of the recognition process. This paper is an extensive review of effective neural network approaches for handwritten character recognition models. The proposed classifier methods in the literature were explored from handwritten text recognition. The challenges prevail in the feature extraction and word combination slows down the scope of classification tasks. From the conducted review, it can be stated that this research domain has to be explored in a fine -grained way of constructing the classifiers.

Published
2021-11-10
How to Cite
Sunitha S Nair, Dr. P. Ranjit Jeba Thangaiah. (2021). A Study on Handwritten Character Recognition Mechanisms using Neural Networks. Design Engineering, 11150 - 11166. Retrieved from http://thedesignengineering.com/index.php/DE/article/view/6176
Section
Articles