A Review on Handwritten Devanagari Character Recognition

EasyChair Preprint 2128, version 2

Abstract

Because of the vast variation in writing styles, the handwritten text recognition is a challenging task. In India, a large number of people use Devanagari Script to write their documents, but due to large complexity, research work accomplished on this script is much lesser as com­pared to English script. Hence, recognition of handwritten Devanagari Script is amongst the most demanding research areas in the field of pattern recognition.  Feature extraction and classification are important steps of OCR which affects the overall accuracy of the character recognition system. This paper gives a comparative study on different techniques used for feature extraction and classification by the researchers over the last few years.

Keywords: OCR, Devanagari, Handwritten character recognition, online, offline, ANN, K-NN, SVM, CNN.

Keyphrases: Artificial Neural Network, CNN, Devanagari, OCR, SVM, k-NN

@booklet{EasyChair:2128,
  author    = {Manoj Sonkusare and Roopam Gupta and Asmita Moghe},
  title     = {A Review on Handwritten Devanagari Character Recognition},
  howpublished = {EasyChair Preprint 2128},
  year      = {EasyChair, 2020}}