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

Automatic target recognition using a feature-decomposition and data-decomposition modular neural network.

L C Wang1, S Z Der, N M Nasrabadi

  • 1SONY Semicond. Co. if America, San Jose, CA 95134, USA. lwang@ssa-de.sel.sony.com

IEEE Transactions on Image Processing : a Publication of the IEEE Signal Processing Society
|February 16, 2008
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

[Progress in the development of pertussis-containing vaccines and immunization strategies].

Zhonghua yu fang yi xue za zhi [Chinese journal of preventive medicine]·2026
Same author

[Influencing factors analysis and prediction model establishment of toe-amputation in patients with diabetic foot].

Zhonghua yi xue za zhi·2025
Same author

[Epidemic characteristics and trend analysis of major injuries deaths among children and adolescents in Jiangsu Province from 2012 to 2021].

Zhonghua liu xing bing xue za zhi = Zhonghua liuxingbingxue zazhi·2024
Same author

[Effect of recombinant human thrombin for hemostasis in liver resection: a randomized controlled phase Ⅲ clinical trial].

Zhonghua yi xue za zhi·2023
Same author

[Analysis of fast-growing culturable bacteria and pathogenic bacteria in the surface water of the northeast coast of Hainan Island in China].

Zhonghua yu fang yi xue za zhi [Chinese journal of preventive medicine]·2023
Same author

[Incidence of gastric cancer and risk factors in Suzhou cohort].

Zhonghua liu xing bing xue za zhi = Zhonghua liuxingbingxue zazhi·2022
Same journal

Style-Aware Contrastive Test-Time Adaptation: A Dual-Cache Model for Robust Vision-Language Alignment.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
Same journal

Semantic Frame Interpolation.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
Same journal

Physics-Guided Cross-Modal Decoupling with Test-Time Adaptation for Hyperspectral Image Restoration.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
Same journal

Change-Prior-Guided Unsupervised Change Detection of Heterogeneous Remote Sensing Images.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
Same journal

AgonicDreamer: Enhancing Multi-View Consistency in Text-to-3D Generation via Rectified Score Distillation.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
Same journal

BiCM-Prompt: Bidirectional Cross-Modal Prompt Tuning for Class-Incremental Learning on Multisource Remote Sensing Images.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
See all related articles

This study introduces a modular neural network for automatic target recognition in forward-looking infrared (FLIR) imagery. Feature decomposition enhances classification performance and reduces network complexity compared to traditional methods.

Area of Science:

  • Computer Science
  • Artificial Intelligence
  • Image Processing

Background:

  • Automatic target recognition (ATR) is crucial for analyzing forward-looking infrared (FLIR) imagery.
  • Traditional neural network approaches can be complex and computationally intensive.
  • Modular approaches offer potential for improved efficiency and performance in ATR tasks.

Purpose of the Study:

  • To develop and evaluate a modular neural network classifier for ATR using FLIR imagery.
  • To investigate the benefits of feature decomposition and stacked generalization for improving classifier performance.
  • To assess the effectiveness of a data-decomposition classifier network.

Main Methods:

  • A modular neural network classifier composed of independently trained subnetworks was designed.

Related Experiment Videos

  • Local features were extracted from specific image portions, and decisions were combined.
  • Multiresolution features and stacked generalization were implemented.
  • A data-decomposition classifier network was also developed and tested.

    • Main Results:

    • Feature decomposition led to superior performance over fully connected networks in classification probability and reduced complexity.
    • The use of multiresolution features and stacked generalization further enhanced classifier accuracy.
    • The data-decomposition classifier network also demonstrated improved performance.
    • Experiments were conducted on a substantial dataset of real FLIR images.

    Conclusions:

    • Modular neural network architectures, particularly with feature and data decomposition, offer significant advantages for FLIR-based ATR.
    • The proposed methods improve classification accuracy and reduce model complexity.
    • Stacked generalization and multiresolution features are effective strategies for enhancing ATR systems.