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

Face recognition using total margin-based adaptive fuzzy support vector machines.

Yi-Hung Liu1, Yen-Ting Chen

  • 1Department of Mechanical Engineering, Chung Yuan Christian University, Chung-Li 32023, Taiwan, ROC. lyh@cycu.edu.tw

IEEE Transactions on Neural Networks
|February 7, 2007
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

Tuning the Magnetic Properties of Heterotrimetallic Co-M-Co (M = Ni and Pd) Chain Complexes via Redox Modulation and Metal Ion Replacement.

Inorganic chemistry·2026
Same author

Estimating cognitive function score from mild cognitive impairment to moderate dementia using a hybrid model combining plasma biomarkers with electroencephalogram signal.

Journal of Alzheimer's disease : JAD·2026
Same author

Pyracylenes: Facile Synthetic Access and Continuous Face-To-Face Antiaromatic π‑Stacking.

JACS Au·2026
Same author

Interlocked Rotaxane Enables TADF with Distinct Excited-State Structural Relaxation.

Journal of the American Chemical Society·2026
Same author

Two-Phase Distributed Genetic-Based Algorithm for Time-Aware Shaper Scheduling in Industrial Sensor Networks.

Sensors (Basel, Switzerland)·2026
Same author

Use of brolucizumab in refractory neovascular age-related macular degeneration: characteristics and biomarkers of intraocular inflammation from an Asian real-world study.

Japanese journal of ophthalmology·2025
Same journal

Universal perceptron and DNA-like learning algorithm for binary neural networks: LSBF and PBF implementations.

IEEE transactions on neural networks·2013
Same journal

Guest editorial: special section on white box nonlinear prediction models.

IEEE transactions on neural networks·2011
Same journal

Data-based fault-tolerant control of high-speed trains with traction/braking notch nonlinearities and actuator failures.

IEEE transactions on neural networks·2011
Same journal

Guest editorial: special section on data-based control, modeling, and optimization.

IEEE transactions on neural networks·2011
Same journal

Neural network-based multiple robot simultaneous localization and mapping.

IEEE transactions on neural networks·2011
Same journal

Data-driven model-free adaptive control for a class of MIMO nonlinear discrete-time systems.

IEEE transactions on neural networks·2011
See all related articles

A novel classifier, total margin-based adaptive fuzzy support vector machines (TAF-SVM), enhances face recognition by addressing overfitting and imbalanced data. TAF-SVM demonstrates superior accuracy and stability compared to traditional support vector machines (SVM).

Area of Science:

  • Computer Science
  • Artificial Intelligence
  • Machine Learning

Background:

  • Support Vector Machines (SVMs) face challenges in face recognition, including overfitting due to outliers and skewed hyperplanes from imbalanced datasets.
  • Existing SVM methods may not adequately handle noise and data distribution issues common in real-world face recognition tasks.

Purpose of the Study:

  • To introduce a new classifier, total margin-based adaptive fuzzy support vector machines (TAF-SVM), designed to overcome limitations of traditional SVMs in face recognition.
  • To improve face recognition accuracy and stability by addressing overfitting, imbalanced data, and generalization error.

Main Methods:

  • Developed TAF-SVM by integrating fuzzification of penalty for outlier handling and a different cost algorithm for imbalanced data.

Related Experiment Videos

  • Replaced the conventional soft margin with a total margin algorithm to achieve a lower generalization error bound.
  • Applied TAF-SVM in both linear and nonlinear cases, utilizing Kernel Fisher's Discriminant Analysis (KFDA) for feature extraction on CYCU and FERET face databases.

    • Main Results:

    • Experimental results on the Chung Yuan Christian University (CYCU) multiview and FERET face databases show TAF-SVM outperforms standard SVM in face recognition accuracy.
    • TAF-SVM achieved smaller error variances across multiple tests, indicating enhanced recognition stability.
    • The proposed method effectively mitigates overfitting and corrects hyperplane skew caused by data imbalances.

    Conclusions:

    • TAF-SVM offers a significant improvement over traditional SVM for face recognition tasks.
    • The adaptive fuzzy approach and total margin algorithm contribute to higher accuracy and robust performance, especially with noisy or imbalanced data.
    • TAF-SVM presents a more stable and accurate solution for reliable face recognition systems.