A Systematic Review of Automation in Handwritten Character Recognition

摘要

Handwritten character recognition has been a problem of interest for the researchers for the last few decades still the problem needs practically implementable and scalable solutions to different applications. One of the main challenges in this field is to recognize the cursive handwritten text. The problem is to align the text, segment individual characters and then recognize the characters. In this paper, we describe the needs, applications, basic terminology and specific research contributions related to handwritten character recognition. We line-out steps involved in the process and also classify various methods related to each step. We also provide advantages and disadvantages of various feature selection and classification methods. We conclude with the analytical comparison of the methods and provide the reader with a systematic review of this area. We also mention the future directions towards an optimal handwritten character recognition method.