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 Video

Updated: Jun 6, 2026

Deep Neural Networks for Image-Based Dietary Assessment
13:19

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

Neural-network classifiers for automatic real-world aerial image recognition.

S Greenberg, H Guterman

    Applied Optics
    |November 25, 2010
    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

    SWI versus GRE-T2*: Assessing cortical superficial siderosis in advanced cerebral amyloid angiopathy.

    Revue neurologique·2023
    Same author

    Lecanemab: Appropriate Use Recommendations.

    The journal of prevention of Alzheimer's disease·2023
    Same author

    Discrepancies between tumor genomic profiling and germline genetic testing.

    ESMO open·2022
    Same author

    Real-world impact of anti-HER2 therapy-related cardiotoxicity in patients with advanced HER2-positive breast cancer.

    Asia-Pacific journal of clinical oncology·2020
    Same author

    Response to omalizumab using patient enrichment criteria from trials of novel biologics in asthma.

    Allergy·2017
    Same author

    Sequential exposure to fibroblast growth factors (FGF) 2, 9 and 18 enhances hMSC chondrogenic differentiation.

    Osteoarthritis and cartilage·2014
    Same journal

    Multifunctional reconfigurable terahertz metasurface based on vanadium dioxide phase transition: achieving broadband absorption and efficient polarization conversion.

    Applied optics·2026
    Same journal

    High-Q-factor electromagnetically induced transparency utilizing quasi-bound states in the continuum in an all-dielectric terahertz metasurface.

    Applied optics·2026
    Same journal

    Automated stitching interferometry for high-precision metrology of X-ray mirrors.

    Applied optics·2026
    Same journal

    Experimental demonstration of an approach to designing a metal-dielectric DBR resonant cavity structure.

    Applied optics·2026
    Same journal

    High-precision wavefront reconstruction from a single-shot interferogram using a physics-driven hybrid feature calibration network.

    Applied optics·2026
    Same journal

    Ultra-high-Q Fano resonance based on coupled topological corner states in Kagome photonic crystals.

    Applied optics·2026
    See all related articles

    This study explores neural networks for automatic aerial image recognition. The multilayer perceptron network excelled at handling combined geometric distortions, outperforming other methods for robust target recognition.

    Area of Science:

    • Computer Vision
    • Artificial Intelligence
    • Pattern Recognition

    Background:

    • Automatic aerial image recognition (AAIR) is crucial for tasks like target tracking.
    • Achieving invariance to position and orientation is a key challenge in AAIR.

    Purpose of the Study:

    • To apply and evaluate multilayer perceptron (MLP) and adaptive resonance theory version 2-A (ART 2-A) networks for AAIR.
    • To assess the effectiveness of invariant feature spaces combined with neural networks for robust image classification.

    Main Methods:

    • Utilized invariant feature spaces including Fourier-transform, Zernike moments, central moments, and polar transforms.
    • Implemented supervised (MLP) and unsupervised (ART 2-A) neural network classifiers.
    • Compared MLP performance against classical correlators, specifically the binary phase-only filter (BPOF) correlator.

    Related Experiment Videos

    Last Updated: Jun 6, 2026

    Deep Neural Networks for Image-Based Dietary Assessment
    13:19

    Deep Neural Networks for Image-Based Dietary Assessment

    Published on: March 13, 2021

    Main Results:

    • The MLP network demonstrated superior performance, outperforming the BPOF correlator.
    • ART 2-A offered speed and required fewer training vectors but lacked robustness to combined distortions.
    • Only the MLP classifier successfully handled combined shift and rotation geometric distortions.

    Conclusions:

    • MLP networks combined with invariant features provide a powerful approach for AAIR.
    • The choice of neural network depends on specific requirements regarding speed, training data, and distortion tolerance.
    • MLP networks are particularly suitable for AAIR tasks demanding resilience to complex geometric transformations.