Jove
Visualize
Contact Us
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

Related Experiment Videos

Style context with second-order statistics.

Sriharsha Veeramachaneni1, George Nagy

  • 1Automated Reasoning Division, Istituto per la Ricerca Scientifica e Tecnologica, Via Sommarive 18, Povo, Trento 38050, Italy. sriharsha@itc.it

IEEE Transactions on Pattern Analysis and Machine Intelligence
|January 5, 2005
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

Comment: projection methods require black border removal.

IEEE transactions on pattern analysis and machine intelligence·2009
Same author

A comparative study of local matching approach for face recognition.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2007
Same author

Analytical results on style-constrained bayesian classification of pattern fields.

IEEE transactions on pattern analysis and machine intelligence·2007
Same author

Style consistent classification of isogenous patterns.

IEEE transactions on pattern analysis and machine intelligence·2005
Same author

Rapid automated three-dimensional tracing of neurons from confocal image stacks.

IEEE transactions on information technology in biomedicine : a publication of the IEEE Engineering in Medicine and Biology Society·2002

This study introduces a style-conscious classifier to improve pattern recognition by modeling statistical dependencies between patterns from the same source. The new method reduces error rates by up to 20% in handwritten digit recognition tasks.

Area of Science:

  • Pattern recognition
  • Machine learning
  • Statistical modeling

Background:

  • Patterns often originate from a common source, creating statistical dependencies known as style context.
  • Multisource recognition problems are challenged by these inherent dependencies.

Purpose of the Study:

  • To model and leverage statistical dependencies (style context) for improved pattern recognition.
  • To develop an optimal classifier for normally distributed styles.

Main Methods:

  • Modeling dependencies using second-order statistics.
  • Formulating an optimal classifier for normally distributed styles.
  • Estimating model parameters from class pairs for versatile classifier training.

Main Results:

Related Experiment Videos

  • Model parameters estimated from class pairs are sufficient for training classifiers across varying test field lengths.
  • The style-conscious classifier achieved up to a 20% reduction in field error rate.
  • Performance was validated on quadruples of handwritten digits from NIST datasets.

Conclusions:

  • Incorporating style context significantly enhances pattern recognition accuracy.
  • The proposed method offers a computationally intensive yet effective approach for multisource recognition.
  • The findings demonstrate the utility of second-order statistics in capturing and utilizing style information.