Recognition of hand-printed characters via Induct-RDR

Author

Amin, Adnan ; Singh, Sameer

Author_Institution

Sch. of Comput. Sci. & Eng., New South Wales Univ., Kensington, NSW, Australia

fYear

1999

fDate

20-22 Sep 1999

Firstpage

221

Lastpage

224

Abstract

The goal of character recognition research is to simplify and automate the development of character recognition algorithms. We describe an approach based on applying preprocessing to data sets of Latin characters and then applying a machine learning approach to the data sets to build a knowledge base able to classify unseen pre-processed characters. The machine learning method, Induct/RDR, has a number of features that make it particularly suitable for character recognition. It has the potential to integrate automatic analysis with a manual knowledge acquisition methodology if further refinement is required. Initial results on hand-printed Latin characters show the recognition accuracy of up to 90.2% on unseen cases for the machine learning system

Keywords

handwriting recognition; handwritten character recognition; knowledge acquisition; learning (artificial intelligence); natural languages; text analysis; Induct-RDR; Latin characters; automatic analysis; character recognition algorithms; character recognition research; data preprocessing; data sets; hand-printed Latin characters; hand-printed character recognition; knowledge base; machine learning approach; machine learning method; machine learning system; manual knowledge acquisition methodology; recognition accuracy; unseen cases; unseen pre-processed characters; Character recognition; Decision support systems;

fLanguage

English

Publisher

ieee

Conference_Titel

Document Analysis and Recognition, 1999. ICDAR '99. Proceedings of the Fifth International Conference on

Conference_Location

Bangalore

Print_ISBN

0-7695-0318-7

Type

conf

DOI

10.1109/ICDAR.1999.791764

Filename

791764