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

Hidden Markov model-based weighted likelihood discriminant for 2-D shape classification.

Ninad Thakoor1, Jean Gao, Sungyong Jung

  • 1Electrical Engineering Department, University of Texas at Arlington, Arlington, TX 76010, USA. ninad.thakoor@uta.edu

IEEE Transactions on Image Processing : a Publication of the IEEE Signal Processing Society
|November 10, 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

DAG-VAERL: a novel causal inference method for building causal gene regulatory networks.

BioData mining·2026
Same author

Transcriptome graph transformer: a graph transformer-based unsupervised model for transcriptome data analysis.

BMC bioinformatics·2026
Same author

Structural and Electronic Engineering of WSe<sub>2</sub> for High-Performance Gas Sensing.

ACS sensors·2026
Same author

Segment Any Cell: A SAM-Based Auto-Prompting Fine-Tuning Framework for Nuclei Segmentation.

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

Abnormality-aware multimodal learning for WSI classification.

Frontiers in medicine·2025
Same author

From correlation to causation using directed topological overlap matrix: Applications in genomics.

Methods (San Diego, Calif.)·2023
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 novel weighted likelihood discriminant for shape classification, improving accuracy by minimizing errors. The method uses hidden Markov models (HMMs) for shape curvature, outperforming traditional maximum likelihood approaches.

Area of Science:

  • Computer Vision
  • Pattern Recognition
  • Machine Learning

Background:

  • Traditional maximum likelihood (ML) methods classify shapes based on independent models, potentially limiting accuracy.
  • General hidden Markov model (HMM) methods also rely on independent class probabilities.
  • Existing shape descriptors like Fourier descriptors and Zernike moments are often used with support vector machines (SVMs).

Purpose of the Study:

  • To present a weighted likelihood discriminant for minimum error shape classification.
  • To develop a novel approach that utilizes information from all classes to minimize classification errors.
  • To compare the proposed method against traditional ML and other classification techniques.

Main Methods:

  • Utilizing a hidden Markov model (HMM) for shape curvature as a 2-D shape descriptor.

Related Experiment Videos

  • Introducing a weighted likelihood discriminant function.
  • Implementing a minimum classification error strategy using a generalized probabilistic descent method.

    • Main Results:

    • The proposed weighted likelihood discriminant method demonstrates improved shape classification accuracy.
    • Comparative analysis shows superior performance over classic ML classification with various HMM topologies.
    • The approach also outperforms Fourier descriptor and Zernike moments-based SVM classification for diverse shapes.

    Conclusions:

    • The weighted likelihood discriminant offers a more effective approach to shape classification by minimizing errors.
    • Leveraging HMMs for shape curvature and incorporating all class information enhances classification performance.
    • This method provides a robust alternative for accurate shape recognition tasks.