Jove
Visualize
Contact Us

Related Experiment Video

Updated: May 29, 2026

Generating Strictly Controlled Stimuli for Figure Recognition Experiments
05:39

Generating Strictly Controlled Stimuli for Figure Recognition Experiments

Published on: March 18, 2019

Reliable indexing using unreliable recognition devices.

J K Mullin1

  • 1Department of Computer Science, University of Western Ontario, London, Ont., Canada.

IEEE Transactions on Pattern Analysis and Machine Intelligence
|August 27, 2011
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Interactive programs for statistical computations and information display in chronic obstructive pulmonary disease research.

Computers and biomedical research, an international journal·1974
Same author

Changes in corticosteroid usage by chronic asthma patients during a year of cromoglycate treatment.

The Journal of allergy and clinical immunology·1973
See all related articles
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

This study introduces a novel method to create reliable document indexes from unreliable character recognition devices. By transforming common errors into pseudocharacters, the approach significantly improves indexing accuracy.

Area of Science:

  • Information Science
  • Computer Science
  • Data Processing

Background:

  • Character recognition devices often produce substitution errors, impacting document indexing reliability.
  • Existing methods struggle to consistently manage these errors, leading to inaccurate information retrieval.

Purpose of the Study:

  • To develop and evaluate a new method for generating reliable document indexes from unreliable character recognition.
  • To address common substitution errors in character recognition through a novel transformation technique.

Main Methods:

  • A pseudocharacter transformation strategy was implemented to consolidate likely substitution errors.
  • The method's precision and recall were analyzed using published error matrices and word frequency data.
  • Evaluation involved a large corpus of documents processed by the unreliable character recognition device.

Related Experiment Videos

Last Updated: May 29, 2026

Generating Strictly Controlled Stimuli for Figure Recognition Experiments
05:39

Generating Strictly Controlled Stimuli for Figure Recognition Experiments

Published on: March 18, 2019

Main Results:

  • The developed method demonstrated surprisingly good performance in producing reliable indexes.
  • Analysis showed effective handling of character confusions, leading to improved indexing accuracy.
  • Precision and recall metrics indicated a significant reduction in indexing errors.

Conclusions:

  • The pseudocharacter transformation method offers a robust solution for improving document indexing with unreliable OCR.
  • The approach provides a foundation for more accurate and reliable information retrieval systems.
  • Further enhancements are suggested to optimize the method's performance and applicability.