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

Two class minimax distance transform correlation filter.

Marios Savvides1, B V K Vijaya Kumar, Pradeep K Khosla

  • 1Department of Electrical and Computer Engineering, Carnegie Mellon University, Pittsburgh, Pennsylvania 15213, USA.

Applied Optics
|November 21, 2002
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

Measuring the potential risk of re-identification of imaging research participants from open-source automated face recognition software.

NeuroImage·2025
Same author

A face-off of MRI research sequences by their need for de-facing.

NeuroImage·2023
Same author

Wastewater sequencing reveals early cryptic SARS-CoV-2 variant transmission.

Nature·2022
Same author

Face recognition from research brain PET: An unexpected PET problem.

NeuroImage·2022
Same author

Wastewater sequencing uncovers early, cryptic SARS-CoV-2 variant transmission.

medRxiv : the preprint server for health sciences·2022
Same author

Mutual Information Regularized Feature-Level Frankenstein for Discriminative Recognition.

IEEE transactions on pattern analysis and machine intelligence·2021

A novel minimax distance transform correlation filter (MDTCF) enhances class separation by minimizing true-class distance and maximizing false-class distance to a reference image. This method shows improved discrimination performance in face recognition tasks.

Area of Science:

  • Computer Vision
  • Machine Learning
  • Pattern Recognition

Background:

  • Correlation filters are widely used in pattern recognition and computer vision tasks.
  • Existing methods like the distance classifier correlation filter (DCCF) have limitations in discriminating between classes.

Purpose of the Study:

  • To introduce a new correlation filter formulation, the minimax distance transform correlation filter (MDTCF).
  • To enhance the separation between true-class and false-class correlation outputs.
  • To evaluate the discrimination performance of MDTCF compared to DCCF.

Main Methods:

  • Developed the minimax distance transform correlation filter (MDTCF) formulation.
  • MDTCF minimizes average squared distance for true-class images and maximizes it for false-class images relative to a reference.

Related Experiment Videos

  • Classification is performed by comparing the squared distance of a filtered test image to the reference image.

    • Main Results:

    • MDTCF formulation demonstrated increased separation between true-class and false-class correlation outputs.
    • MDTCF was shown to be a generalization of the distance classifier correlation filter (DCCF).
    • MDTCF exhibited improved discrimination performance over DCCF.

    Conclusions:

    • The proposed MDTCF offers enhanced discrimination capabilities for classification tasks.
    • MDTCF provides a more robust approach to distinguishing between similar classes in pattern recognition.
    • Performance evaluation on face databases confirmed the effectiveness of MDTCF.