Handwritten Mathematical Recognition Tool

Abstract

The recognition of handwritten mathematical expressions has received an increasing amount of attention in pattern recognition research. It is the process of taking in raw data and making actions based on the category of the data. In this paper, we present a tool for recognizing handwritten mathematical expressions. The proposed architecture aims at handling the handwritten expressions by performing segmentation of the input based on each pen ups and pen downs followed by symbol classification. As a classifier, an Extreme Learning Machine and Support Vector machines are used, the classifier which produces a best accuracy is selected and then the symbols are trained among various handwritten mathematical expression and a promising result are achieved at symbol classification stage. Once the symbols are classified, the corresponding output is converted to LaTex format.