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

Affine invariant pattern recognition using Multiscale Autoconvolution.

Esa Rahtu1, Mikko Salo, Janne Heikkilä

  • 1Machine Vision Group, Infotech Oulu, Finland. erahtu@ee.oulu.fi

IEEE Transactions on Pattern Analysis and Machine Intelligence
|June 10, 2005
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

KaSARi: a national framework for standardized, automated, and predictive radiotherapy in Finland.

Acta oncologica (Stockholm, Sweden)·2026
Same author

Prompt is All You Need: Prompting Foundation Models for Large-Scale Self-Supervised Semantic Segmentation.

IEEE transactions on pattern analysis and machine intelligence·2026
Same author

Automated daily adaptation for breast cancer radiotherapy: benefits of deep learning-driven dose prediction workflow for CBCT-based adaptation.

Physics in medicine and biology·2026
Same author

Association of childhood adversities on the persistence and remission of drug-naïve attention-deficit/hyperactivity disorder in a longitudinal community-based sample.

Psychiatry research·2025
Same author

Clinical outcome of high-dose-rate interstitial brachytherapy in vulvar cancer: A single institutional experience.

Brachytherapy·2025
Same author

A comprehensive dose-volume histogram-based index for radiotherapy treatment plan quality evaluation: application to breast cancer radiotherapy.

Physics in medicine and biology·2025
Same journal

HardFlow: Hard-Constrained Sampling for Flow-Matching Models Via Trajectory Optimization.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Industrial Brain: Self-Evolving Neuro-Symbolic Autonomy with Causal Resilience for Cyber-Physical Systems.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Adaptive Hardness-Driven Dictionary Distillation for Incomplete Streaming View Clustering.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Mixture of Global and Local Experts with Diffusion Transformer for Controllable Face Generation.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Task-KV: Task-aware KV Cache Optimization via Semantic Differentiation of Attention Heads.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Achieving Text-based Person Retrieval with Any Granularity.

IEEE transactions on pattern analysis and machine intelligence·2026
See all related articles

A new Multiscale Autoconvolution (MSA) transform offers affine invariant image analysis. This method efficiently classifies objects even with significant image distortions, outperforming other techniques.

Area of Science:

  • Computer Vision
  • Image Processing
  • Pattern Recognition

Background:

  • Affine transformations are common in image distortions.
  • Existing invariant transforms often require complex preprocessing like boundary or interest point extraction.
  • Computational efficiency is crucial for real-time applications.

Purpose of the Study:

  • Introduce a novel affine invariant image transform: Multiscale Autoconvolution (MSA).
  • Develop a method for robust object classification under affine distortions.
  • Reduce computational complexity compared to existing techniques.

Main Methods:

  • The Multiscale Autoconvolution (MSA) transform is based on a probabilistic image function interpretation.
  • Utilizes the Fast Fourier Transform (FFT) for significant computational load reduction.

Related Experiment Videos

  • Applies transform values directly as descriptors for pattern classification.

    • Main Results:

    • Demonstrated the effectiveness of MSA transform values in various object classification tasks.
    • Achieved affine invariance without explicit boundary or interest point extraction.
    • Showcased computational efficiency through FFT implementation.

    Conclusions:

    • The Multiscale Autoconvolution (MSA) transform is a viable method for affine invariant pattern classification.
    • The technique is particularly suitable for image analysis where distortions can be modeled as affine transformations.
    • Offers a computationally efficient alternative to existing invariant methods.