Accuracy improvement of handwritten numeral recognition by mirror image learning

Tetsushi Wakabayashi, Meng Shi, Wataru Ohyama, Fumitaka Kimura

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

This paper proposes a new corrective learning algorithm and evaluates the performance by handwritten numeral recognition test. The algorithm generates a mirror image of a pattern that belongs to one class of a pair of confusing classes and utilizes it as a learning pattern of the other class. This paper also studies on how to extract confusing patterns within a certain margin of a decision boundary to generate enough number of mirror images, and how to perform an effective mirror image compensation to increase the margin. Recognition accuracies of the minimum distance classifier and the projection distance method were improved from 93.17% to 98.38% and from 99.11% to 99.41% respectively in the recognition test for handwritten numeral database IPTP CD-ROMI.

Original languageEnglish
Title of host publicationProceedings - 6th International Conference on Document Analysis and Recognition, ICDAR 2001
PublisherIEEE Computer Society
Pages338-343
Number of pages6
ISBN (Electronic)0769512631, 0769512631, 0769512631
DOIs
Publication statusPublished - 2001
Event6th International Conference on Document Analysis and Recognition, ICDAR 2001 - Seattle, United States
Duration: Sept 10 2001Sept 13 2001

Publication series

NameProceedings of the International Conference on Document Analysis and Recognition, ICDAR
Volume2001-January
ISSN (Print)1520-5363

Other

Other6th International Conference on Document Analysis and Recognition, ICDAR 2001
Country/TerritoryUnited States
CitySeattle
Period9/10/019/13/01

All Science Journal Classification (ASJC) codes

  • Computer Vision and Pattern Recognition

Fingerprint

Dive into the research topics of 'Accuracy improvement of handwritten numeral recognition by mirror image learning'. Together they form a unique fingerprint.

Cite this