A comparative study of different feature sets for recognition of handwritten Arabic numerals using a Multi Layer Perceptron

TL;DR

This study compares seven feature sets for recognizing handwritten Arabic numerals with an MLP classifier, identifying effective combinations that achieve up to 95.80% accuracy, aiding OCR applications.

Contribution

It provides a comparative analysis of various feature sets for handwritten numeral recognition and identifies effective feature combinations for improved accuracy.

Findings

Maximum recognition rate of 95.80% achieved.

Shadow and centroid features combination is most effective.

Smaller feature set performs well for OCR applications.

Abstract

The work presents a comparative assessment of seven different feature sets for recognition of handwritten Arabic numerals using a Multi Layer Perceptron (MLP) based classifier. The seven feature sets employed here consist of shadow features, octant centroids, longest runs, angular distances, effective spans, dynamic centers of gravity, and some of their combinations. On experimentation with a database of 3000 samples, the maximum recognition rate of 95.80% is observed with both of two separate combinations of features. One of these combinations consists of shadow and centriod features, i. e. 88 features in all, and the other shadow, centroid and longest run features, i. e. 124 features in all. Out of these two, the former combination having a smaller number of features is finally considered effective for applications related to Optical Character Recognition (OCR) of handwritten Arabic numerals. The work can also be extended to include OCR of handwritten characters of Arabic alphabet.