Artificial Neural Network Based Optical Character Recognition

TL;DR

This paper presents an OCR method that uses feature extraction based on topological and geometrical properties, combined with an artificial neural network for character recognition, inspired by human shape perception.

Contribution

It introduces a feature-based neural network approach for OCR that emphasizes shape and property extraction, enhancing recognition accuracy.

Findings

Effective recognition of characters using feature vectors

Improved classification accuracy with neural network approach

Utilizes shape-based features inspired by human perception

Abstract

Optical Character Recognition deals in recognition and classification of characters from an image. For the recognition to be accurate, certain topological and geometrical properties are calculated, based on which a character is classified and recognized. Also, the Human psychology perceives characters by its overall shape and features such as strokes, curves, protrusions, enclosures etc. These properties, also called Features are extracted from the image by means of spatial pixel-based calculation. A collection of such features, called Vectors, help in defining a character uniquely, by means of an Artificial Neural Network that uses these Feature Vectors.