Algorithms for Recognizing Contour-Traced Handprinted Characters

Godfried T. Toussaint; Robert W. Donaldson

doi:10.1109/T-C.1970.222972

Algorithms for Recognizing Contour-Traced Handprinted Characters

Godfried T. Toussaint, Robert W. Donaldson

Computer Science

Research output: Contribution to journal › Article › peer-review

Abstract

A contour-tracing technique originally devised by Clemens and Mason was modified and used with several different classifiers to recognize upper case handprinted characters. Recognition accuracies obtained compare favorably with other published results, particularly when additional simple tests are performed to differentiate commonly confused characters. One suboptimum classifier, in addition to yielding near optimum performance when tested on training data, uses much less statistical information than the optimum Bayes classifier and is significantly better than the optimum classifier when training and test data are limited.

Original language	English (US)
Pages (from-to)	541-546
Number of pages	6
Journal	IEEE Transactions on Computers
Volume	C-19
Issue number	6
DOIs	https://doi.org/10.1109/T-C.1970.222972
State	Published - Jun 1970

Keywords

Character recognition
classification algorithms
contextual constraints
contour analysis
decision trees
feature extraction
handprinted characters
limited-data experiments
machine learning
suboptimum classification

ASJC Scopus subject areas

Software
Theoretical Computer Science
Hardware and Architecture
Computational Theory and Mathematics

Access to Document

10.1109/T-C.1970.222972

Cite this

@article{96d20166602b48d6beea732ec2c513e4,

title = "Algorithms for Recognizing Contour-Traced Handprinted Characters",

abstract = "A contour-tracing technique originally devised by Clemens and Mason was modified and used with several different classifiers to recognize upper case handprinted characters. Recognition accuracies obtained compare favorably with other published results, particularly when additional simple tests are performed to differentiate commonly confused characters. One suboptimum classifier, in addition to yielding near optimum performance when tested on training data, uses much less statistical information than the optimum Bayes classifier and is significantly better than the optimum classifier when training and test data are limited.",

keywords = "Character recognition, classification algorithms, contextual constraints, contour analysis, decision trees, feature extraction, handprinted characters, limited-data experiments, machine learning, suboptimum classification",

author = "Toussaint, {Godfried T.} and Donaldson, {Robert W.}",

year = "1970",

month = jun,

doi = "10.1109/T-C.1970.222972",

language = "English (US)",

volume = "C-19",

pages = "541--546",

journal = "IEEE Transactions on Computers",

issn = "0018-9340",

publisher = "IEEE Computer Society",

number = "6",

}

TY - JOUR

T1 - Algorithms for Recognizing Contour-Traced Handprinted Characters

AU - Toussaint, Godfried T.

AU - Donaldson, Robert W.

PY - 1970/6

Y1 - 1970/6

N2 - A contour-tracing technique originally devised by Clemens and Mason was modified and used with several different classifiers to recognize upper case handprinted characters. Recognition accuracies obtained compare favorably with other published results, particularly when additional simple tests are performed to differentiate commonly confused characters. One suboptimum classifier, in addition to yielding near optimum performance when tested on training data, uses much less statistical information than the optimum Bayes classifier and is significantly better than the optimum classifier when training and test data are limited.

AB - A contour-tracing technique originally devised by Clemens and Mason was modified and used with several different classifiers to recognize upper case handprinted characters. Recognition accuracies obtained compare favorably with other published results, particularly when additional simple tests are performed to differentiate commonly confused characters. One suboptimum classifier, in addition to yielding near optimum performance when tested on training data, uses much less statistical information than the optimum Bayes classifier and is significantly better than the optimum classifier when training and test data are limited.

KW - Character recognition

KW - classification algorithms

KW - contextual constraints

KW - contour analysis

KW - decision trees

KW - feature extraction

KW - handprinted characters

KW - limited-data experiments

KW - machine learning

KW - suboptimum classification

UR - http://www.scopus.com/inward/record.url?scp=0014796565&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0014796565&partnerID=8YFLogxK

U2 - 10.1109/T-C.1970.222972

DO - 10.1109/T-C.1970.222972

M3 - Article

AN - SCOPUS:0014796565

SN - 0018-9340

VL - C-19

SP - 541

EP - 546

JO - IEEE Transactions on Computers

JF - IEEE Transactions on Computers

IS - 6

ER -

Algorithms for Recognizing Contour-Traced Handprinted Characters

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this