Artificial neural networks and fuzzy logic for recognizing alphabet characters and mathematical symbols

TL;DR

This paper introduces a novel OCR approach combining artificial neural networks with fuzzy logic-based segmentation to improve recognition of alphabet characters and mathematical symbols.

Contribution

It presents an original enhancement of the backpropagation algorithm and a fuzzy logic-based image segmentation method for recognizing touching characters.

Findings

Improved recognition accuracy for characters and symbols

Effective segmentation of touching characters using fuzzy logic

Enhanced pattern recognition performance

Abstract

Optical Character Recognition software (OCR) are important tools for obtaining accessible texts. We propose the use of artificial neural networks (ANN) in order to develop pattern recognition algorithms capable of recognizing both normal texts and formulae. We present an original improvement of the backpropagation algorithm. Moreover, we describe a novel image segmentation algorithm that exploits fuzzy logic for separating touching characters.