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

Representing images using nonorthogonal Haar-like bases.

Feng Tang1, Ryan Crabb, Hai Tao

  • 1Department of Computer Engineering, Univesity of California, Santa Cruz, CA 95064, USA. tang@soe.ucsc.edu

IEEE Transactions on Pattern Analysis and Machine Intelligence
|October 16, 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

A putative triradial macrofossil from the Ediacaran Jiangchuan Biota.

iScience·2024
Same author

A Bioengineered Nanovesicle Vaccine Boosts T-B cell Interaction for Immunotherapy of Echinococcus multilocularis.

Angewandte Chemie (International ed. in English)·2024
Same author

<i>PtoMYB031</i>, the R2R3 MYB transcription factor involved in secondary cell wall biosynthesis in poplar.

Frontiers in plant science·2024
Same author

[Driving Forces and Mitigation Potential of CO<sub>2</sub> Emissions for Ship Transportation in Guangdong Province, China].

Huan jing ke xue= Huanjing kexue·2024
Same author

PLA-Based Composite Panels Prepared via Multi-Material Fused Filament Fabrication and Associated Investigation of Process Parameters on Flexural Properties of the Fabricated Composite.

Polymers·2024
Same author

Extraction of Polysaccharides from Root of <i>Pseudostellaria heterophylla</i> (Miq.) Pax. and the Effects of Ultrasound Treatment on Its Properties and Antioxidant and Immune Activities.

Molecules (Basel, Switzerland)·2024

This study introduces novel methods using Haar-like binary box functions for efficient image representation. These techniques, including non-orthogonal binary subspace and binary principal component analysis, significantly improve image matching and recognition speed.

Area of Science:

  • Computer Vision
  • Image Processing
  • Machine Learning

Background:

  • Efficient and compact image representation is crucial in computer vision.
  • Existing methods may lack computational efficiency or fail to capture data structure effectively.

Purpose of the Study:

  • To propose novel subspace methods for image representation using Haar-like binary box functions.
  • To develop computationally efficient algorithms for image analysis tasks.

Main Methods:

  • Utilized Haar-like binary box functions for image representation.
  • Introduced the non-orthogonal binary subspace (NBS) method.
  • Developed the binary principal component analysis (B-PCA) algorithm, guided by PCA on NBS.

Main Results:

Related Experiment Videos

  • NBS enables fast template matching and image representation.
  • B-PCA combines computational speed with data structure modeling.
  • BPCA base vectors demonstrated near-orthogonality, allowing approximation of pseudo-inverse projection by dot product.
  • Experiments showed promising performance in image matching, reconstruction, and recognition with significant speedups.

Conclusions:

  • The proposed NBS and B-PCA methods offer efficient and effective solutions for image representation.
  • These methods provide significant computational advantages for various computer vision applications.
  • The novel B-PCA algorithm enhances image analysis by balancing speed and data structure representation.