Handwritten Bangla Digit Recognition Using Chemical Reaction Optimization

EasyChair Preprint 291

Abstract

Handwritten character recognition, an active research field of Artificial Intelligence and Pattern Recognition, has gained enormous attention in recent years. However, much of the work were concentrated on recognition of few economically important languages; limited attention had been paid for character recognition of Bengali, the 5th ranked language of the world. In this paper, we propose a new methodology to recognize handwritten Bengali numerals using a recently established optimization algorithm known as chemical reaction optimization (CRO) in order to increase the recognition accuracy. The proposed method produces a higher accuracy rate, 98.96% which is higher than the outcome of any other proposed method. Clearly, the result suggests that our proposed technique outperforms all previously developed methodologies.

Keyphrases: Chemical Reaction Optimization Algorithm, Handwritten Bangla digit recognition, Operator design, Support Vector Machine, feature extraction

@booklet{EasyChair:291,
  author    = {Pritam Khan Boni and Bappy Shahriar Abir and H.M. Mehedi Hasan and Md. Rafiqul Islam},
  title     = {Handwritten Bangla Digit Recognition Using Chemical Reaction Optimization},
  doi       = {10.29007/1w6r},
  howpublished = {EasyChair Preprint 291},
  year      = {EasyChair, 2018}}