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

Fusing images with different focuses using support vector machines.

Shutao Li1, James Tin-Yau Kwok, Ivor Wai-Hung Tsang

  • 1College of Electrical and Information Engineering, Hunan University, 410082 Changsha, PROC. shutao_li@yahoo.com.cn

IEEE Transactions on Neural Networks
|November 30, 2004
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

ASSR-Net: Anisotropic Structure-Aware and Spectrally Recalibrated Network for Hyperspectral Image Fusion.

IEEE transactions on neural networks and learning systems·2026
Same author

Deep Error-Aware Iterative Optimization Network for Broadband Mosaiced Hyperspectral Imaging.

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

Orientation-Guided Homography for Fine-Grained Cross-View Localization.

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

A novel non-competitive p97/VCP inhibitor induces apoptosis and autophagy to suppress colorectal cancer growth.

European journal of medicinal chemistry·2026
Same author

GBNet: Gated Boundary-Aware Network for Camouflaged Object Detection.

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

Spectral-Spatial Dynamic Scan Mamba for Multi-Source Remote Sensing Data Classification.

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

This study introduces an improved image fusion method using discrete wavelet frame transform (DWFT) and support vector machines (SVM) for creating sharper, in-focus images from multiple captures. The new approach enhances focus and outperforms traditional wavelet-based techniques.

Area of Science:

  • Computer Vision
  • Image Processing
  • Signal Processing

Background:

  • Achieving an everywhere-in-focus image is crucial for many vision tasks but limited by lens depth of field.
  • Traditional wavelet-based image fusion methods using discrete wavelet transform (DWT) are common but have limitations.

Purpose of the Study:

  • To enhance image fusion techniques for generating all-in-focus images.
  • To improve upon existing wavelet-based fusion by incorporating translation invariance and advanced feature selection.

Main Methods:

  • Utilized the discrete wavelet frame transform (DWFT) for a translation-invariant signal representation.
  • Employed support vector machines (SVM) trained on DWFT coefficients to select the best focus at each pixel.
  • Developed a composite wavelet representation for the fused image.

Related Experiment Videos

Main Results:

  • The proposed DWFT and SVM-based method demonstrated superior performance compared to traditional DWT-based fusion.
  • Improvements were observed both in visual quality and quantitative metrics.
  • The method effectively selects the most in-focus regions from source images.

Conclusions:

  • The integration of DWFT and SVM offers a significant advancement in image fusion for all-in-focus image generation.
  • This approach provides a more robust and accurate method for recovering sharp images from a series of out-of-focus or varying-focus source images.
  • The proposed technique is a valuable tool for applications requiring high-quality, consistently focused imagery.