Benchmark Classification of Handwritten Dataset by New Operator

Abstract

In recent years, many new classifiers and feature extraction algorithms were proposed and tested on various OCR databases and these techniques were used in wide applications. Various systematic papers and inventions in OCR were reported in the literature. We can say that OCR is one of the most important and active research areas in the pattern recognition. Today, research OCR is dealing with diverse a character of complex problems. Important research in OCR includes the text degraded (heavy noise) and analysis/recognition of complex documents (including texts, images, graphs, tables and video documents). In this proposed system we are suing a new operator Recognition of Devnagari handwritten Characters one of the biggest problem in present scenario. Devnagari characters are not recognized efficiently and truthfully by electronic device. Many researchers and algorithm have been proposed for recognizing of characters. For recognizing of characters, many processes have to be performed but no single technique or algorithm can perform that recognition and give more accurate result. objective of this dissertation work is to propose a new operator, the name of this operator is Kirsch Operator and algorithm for getting accurate result.