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

Facial recognition using multisensor images based on localized kernel eigen spaces.

Satyanadh Gundimada1, Vijayan K Asari

  • 1Symetix, Walla Walla, WA 99362, USA. sgundimada@key.net

IEEE Transactions on Image Processing : a Publication of the IEEE Signal Processing Society
|April 16, 2009
PubMed
Summary

This study introduces a new feature selection and information fusion method to enhance facial recognition accuracy using both visual and thermal images. The novel approach significantly improves performance, overcoming challenges like varying lighting and expressions.

Related Concept Videos

Association Areas of the Cortex01:21

Association Areas of the Cortex

Association areas are regions of the cerebral cortex that do not have a specific sensory or motor function. Instead, they integrate and interpret information from various sources to enable higher cognitive processes such as memory, learning, and decision-making. Some key association areas include the following:
Prefrontal Association Area: This area is located in the frontal lobe and is involved in planning, decision-making, and moderating social behavior. It connects with primary motor areas,...

You might also read

Related Articles

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

Sort by
Same author

Extrapolation Convolution for Data Prediction on a 2-D Grid: Bridging Spatial and Frequency Domains With Applications in Image Outpainting and Compressed Sensing.

IEEE transactions on neural networks and learning systems·2025
Same author

Microscopic nuclei classification, segmentation, and detection with improved deep convolutional neural networks (DCNN).

Diagnostic pathology·2022
Same author

A Deep Neural Network for Early Detection and Prediction of Chronic Kidney Disease.

Diagnostics (Basel, Switzerland)·2022
Same author

Conditional generative adversarial network demosaicing strategy for division of focal plane polarimeters.

Optics express·2020
Same author

Recurrent residual U-Net for medical image segmentation.

Journal of medical imaging (Bellingham, Wash.)·2019
Same author

Breast Cancer Classification from Histopathological Images with Inception Recurrent Residual Convolutional Neural Network.

Journal of digital imaging·2019

Area of Science:

  • Computer Vision
  • Biometrics
  • Machine Learning

Background:

  • Facial recognition systems often struggle with variations in illumination, occlusion, expressions, and temperature.
  • Integrating visual and thermal imaging offers complementary information but requires effective feature extraction and fusion.

Purpose of the Study:

  • To develop and evaluate a novel feature selection and information fusion technique for improving visual and thermal image-based facial recognition.
  • To address common challenges in facial recognition, including illumination, occlusion, expression, and temperature variations.

Main Methods:

  • A modular kernel eigenspaces approach was applied to phase congruency feature maps from individual visual and thermal images.
  • Features were extracted from merged sub-regions and projected into higher dimensional spaces using kernel methods.

Related Experiment Videos

  • A decision-level fusion methodology was combined with the feature selection procedure.

    • Main Results:

    • The proposed localized nonlinear feature selection procedure significantly improved recognition accuracy for both visual and thermal images compared to conventional methods.
    • The combined feature selection and decision-level fusion approach outperformed existing facial recognition techniques.
    • Experimental validation was performed using the AR and Equinox facial databases.

    Conclusions:

    • The developed technique effectively enhances facial recognition accuracy by robustly handling variations.
    • Combining advanced feature selection with decision-level fusion provides a superior approach for multimodal facial recognition systems.