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

Parallel Processing01:20

Parallel Processing

227
The brain processes sensory information rapidly due to parallel processing, which involves sending data across multiple neural pathways at the same time. This method allows the brain to manage various sensory qualities, such as shapes, colors, movements, and locations, all concurrently. For instance, when observing a forest landscape, the brain simultaneously processes the movement of leaves, the shapes of trees, the depth between them, and the various shades of green. This enables a quick and...
227
Visual System01:26

Visual System

689
Light enters the eye through the cornea, a transparent, dome-shaped surface covering the surface of the eyeball that helps to direct and focus incoming light. This light is then channeled toward the pupil, an adjustable opening whose size is controlled by the iris. The iris, a pigmented muscle, regulates the amount of light entering the eye by contracting or dilating the pupil, thereby ensuring optimal light levels for clear vision.
Once through the pupil, the light passes through the lens, a...
689

You might also read

Related Articles

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

Sort by
Same author

Haplotype-resolved T2T genome and population resequencing provide insights into the domestication and mogroside biosynthesis of <i>Siraitia grosvenorii</i> (Cucurbitaceae).

Horticulture research·2026
Same author

Acute low-dose benzo[a]pyrene exposure induces neurodegenerative syndrome in cyp1a low-expression zebrafish: A novel living Bioindicator for dioxin-like environmental pollutants.

Aquatic toxicology (Amsterdam, Netherlands)·2026
Same author

Low-spin state design of highly active diatomic catalysts for oxygen reduction reaction.

National science review·2026
Same author

Chronic heat exposure disrupts hepatic homeostasis in bullfrog (Aquarana catesbeiana) during metamorphosis: Integrated histopathological, biochemical and transcriptomic evidence.

Journal of thermal biology·2025
Same author

An ammonia-induced universal synthesis approach for manganese based layered oxides.

Nature communications·2025
Same author

Chloroplast CASEIN KINASE 2 enhances maize thermotolerance via sHSP26 phosphorylation and retrograde HSFA3 regulation.

Plant physiology·2025

Related Experiment Video

Updated: Sep 12, 2025

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

636

Multi-scale feature pyramid network with bidirectional attention for efficient mural image classification.

Shulan Wang1, Siyu Liu1, Mengting Jin2

  • 1School of Architecture and Art Design, Hebei University of Technology, Tianjin, China.

Plos One
|August 4, 2025
PubMed
Summary
This summary is machine-generated.

This study introduces an advanced deep learning model for mural image recognition, improving accuracy and detail perception for cultural heritage preservation. The model achieves high accuracy and real-time performance, offering a cost-effective digitization solution.

More Related Videos

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

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

501
A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
04:23

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

Published on: April 21, 2023

2.0K

Related Experiment Videos

Last Updated: Sep 12, 2025

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

636
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

501
A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
04:23

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

Published on: April 21, 2023

2.0K

Area of Science:

  • Computer Vision
  • Artificial Intelligence
  • Digital Heritage Preservation

Background:

  • Mural image recognition is crucial for cultural heritage preservation but faces challenges like style generalization, limited data, and image degradation.
  • Existing methods struggle with intricate details and cross-cultural, multi-period style variations.

Purpose of the Study:

  • To develop a robust deep learning model for accurate mural image recognition and digital preservation.
  • To address challenges in generalization, detail perception, and limited sample sizes in mural datasets.

Main Methods:

  • A DenseNet201-FPN model integrated with a Bidirectional Convolutional Block Attention Module (Bi-CBAM).
  • Incorporation of dynamic focal distillation loss and convex regularization for improved performance.
  • Utilized a dynamic temperature distillation strategy for balancing teacher and ground truth supervision.

Main Results:

  • Achieved 87.9% accuracy on a self-constructed mural dataset, a 3.7% improvement over DenseNet201.
  • Enhanced F1-score for rare classes by 6.1% through dynamic distillation.
  • Demonstrated real-time inference capabilities (63ms/frame) on edge devices.

Conclusions:

  • The proposed deep learning model significantly enhances mural image recognition accuracy and detail perception.
  • This approach offers a cost-effective solution for large-scale mural digitization, especially in resource-constrained settings.
  • The model's efficiency and accuracy contribute to advancing digital preservation of cultural heritage.