Position Invariant Optical Character Recognition through Symmetry Features

Author

Holland, Sam ; Neville, Richard

Author_Institution

Machine Learning & Optimisation Res. Group, Univ. of Manchester, Manchester, UK

fYear

2009

fDate

28-29 Dec. 2009

Firstpage

313

Lastpage

315

Abstract

We propose an effective method to achieve position invariance in the application of optical character recognition (OCR). We normalise the position of all inputs based on their symmetry features. The generalized symmetry transform (GST) is used to determine the symmetry features prior to classification by a probabilistic neural network (PNN). We used the United States Postal Service (USPS) data set to measure performance.

Keywords

character recognition; neural nets; transforms; United States Postal Service data set; generalized symmetry transform; position invariant optical character recognition; probabilistic neural network; symmetry features; Character recognition; Neural networks; Neurons; Optical character recognition software; Optical computing; Optical control; Optical network units; Optical reflection; Telecommunication computing; Telecommunication control; character; generalized; invariance; network; neural; position; probabilistic; recognition; symmetry; transform;

fLanguage

English

Publisher

ieee

Conference_Titel

Advances in Computing, Control, & Telecommunication Technologies, 2009. ACT '09. International Conference on

Conference_Location

Trivandrum, Kerala

Print_ISBN

978-1-4244-5321-4

Electronic_ISBN

978-0-7695-3915-7

Type

conf

DOI

10.1109/ACT.2009.84

Filename

5376673