Showing 1–1 of 1 results for author: Armano, T

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

Authors: Giuseppe Airò Farulla, Tiziana Armano, Anna Capietto, Nadir Murru, Rosaria Rossini

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… ▽ More 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. △ Less

Submitted 6 July, 2016; originally announced July 2016.

Journal ref: Lecture Notes in Computer Science, Volume 9759 2016, Computers Helping People with Special Needs, p. 7-14