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 Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

DeepER-Med: Advancing Deep Evidence-Based Research in Medicine Through Agentic AI.

ArXiv·2026
Same author

Toward Practical Solid-State Lithium Batteries With High-Nickel Cathodes: An Interface-Centered Perspective.

Advanced materials (Deerfield Beach, Fla.)·2026
Same author

Generative Artificial Intelligence and Large Language Models in Clinical Oncology.

MedComm·2026
Same author

Investigation on the unsteady aerodynamic coefficients of iced conductors and the applicability of quasi-static assumptions.

Scientific reports·2026
Same author

GPR15-guided CD8<sup>+</sup> T regulatory cells control intestinal inflammation.

Nature·2026
Same author

Self-Limiting Covalent Ligation Mechanism Enabling Anomalously High Interfacial Compatibility in Organic-in-Sulfide All-Solid-State Lithium Batteries.

Angewandte Chemie (International ed. in English)·2026

Related Experiment Video

Updated: Jun 21, 2025

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

2.7K

Deep Learning-Based Dynamic Region of Interest Autofocus Method for Grayscale Image.

Yao Wang1,2, Chuan Wu1, Yunlong Gao1

  • 1Changchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Sciences, Changchun 130033, China.

Sensors (Basel, Switzerland)
|July 13, 2024
PubMed
Summary

This study introduces a novel neural network autofocus method that dynamically selects regions of interest (ROI) for improved optical system focusing. The new approach overcomes limitations of traditional methods, enabling faster and more accurate autofocusing in various scenarios.

Keywords:
autofocusdatasetdeep learninglightweight networkordinal regression

More Related Videos

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

515
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

385

Related Experiment Videos

Last Updated: Jun 21, 2025

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

2.7K
Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

515
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

385

Area of Science:

  • Computer Vision
  • Optical Systems Engineering
  • Machine Learning

Background:

  • Passive autofocus methods, while cost-effective, suffer from focusing failures due to fixed parameters.
  • Limited datasets hinder the application of deep learning in autofocus research.
  • Dynamic region selection is crucial for robust autofocus in complex optical systems.

Purpose of the Study:

  • To develop a deep learning-based autofocus method capable of dynamically selecting regions of interest (ROI).
  • To address the limitations of traditional autofocusing techniques in optical systems.
  • To create a comprehensive dataset for training and evaluating autofocus algorithms.

Main Methods:

  • Constructed a grayscale image dataset for automatic focusing.
  • Formulated autofocus as an ordinal regression problem with two strategies: full-stack search and single-frame prediction.
  • Developed a MobileViT network incorporating a linear self-attention mechanism for dynamic ROI autofocusing.

Main Results:

  • Achieved a Mean Absolute Error (MAE) as low as 0.094 for full-stack search autofocusing.
  • Achieved a MAE of 0.142 for single-frame prediction autofocusing.
  • Demonstrated autofocusing times of 27.8 ms (full-stack) and 27.5 ms (single-frame), showcasing high efficiency.

Conclusions:

  • The proposed neural network autofocus method effectively addresses limitations of passive focusing.
  • Dynamic ROI selection and ordinal regression strategies significantly improve autofocus performance.
  • The MobileViT-based approach offers a promising solution for accurate and rapid autofocus in optical systems.