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

Organoids in cancer therapy: translational applications and clinical promise.

Molecular cancer·2026
Same author

Effects of vitamin D supplementation on patients with systemic lupus erythematosus: A systematic review and meta-analysis.

Autoimmunity reviews·2026
Same author

Electron Beam-Irradiated g-C<sub>3</sub>N<sub>4</sub>/Ti<sub>3</sub>C<sub>2</sub> Nanocomposite Embedded in PVA/SA Hydrogel: An Integrated Platform with Enhanced Photocatalytic Antibacterial Activity.

Gels (Basel, Switzerland)·2026
Same author

Heterogeneity of Obesity and Its Predictive Factors: A Narrative Review of Recent Human and Rodent Studies.

Obesity reviews : an official journal of the International Association for the Study of Obesity·2025
Same author

Recent progress in the pharmacotherapy for obesity.

European journal of pharmacology·2025
Same author

Trends and analysis of risk factor differences in the global burden of chronic kidney disease due to type 2 diabetes from 1990 to 2021: A population-based study.

Diabetes, obesity & metabolism·2025

Related Experiment Video

Updated: Jul 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

2.8K

Element detection and segmentation of mathematical function graphs based on improved Mask R-CNN.

Jiale Lu1, Jianjun Chen1, Taihua Xu1

  • 1School of Computer Science, Jiangsu University of Science and Technology, Zhenjiang 212100, Jiangsu, China.

Mathematical Biosciences and Engineering : MBE
|July 28, 2023
PubMed
Summary

This study introduces an improved Mask R-CNN model for converting mathematical function graphs into tactile graphics, enhancing accessibility for the visually impaired. The new model significantly improves the accuracy of detecting and segmenting graph elements, aiding comprehension of scientific content.

Keywords:
Mask R-CNNattention mechanismfunction graphsinstance segmentationtactile graphics

More Related Videos

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
14:08

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Published on: April 13, 2013

42.7K
Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application
05:56

Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application

Published on: April 14, 2023

2.5K

Related Experiment Videos

Last Updated: Jul 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

2.8K
Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
14:08

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Published on: April 13, 2013

42.7K
Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application
05:56

Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application

Published on: April 14, 2023

2.5K

Area of Science:

  • Computer Vision
  • Assistive Technology
  • Scientific Visualization

Background:

  • Over 2.2 billion people worldwide experience visual impairments, with many relying on non-visual senses.
  • Existing tactile graphics and reading materials for the visually impaired inadequately address the need for graphical content comprehension.
  • Current tactile graphic translation methods face limitations due to image diversity and recognition technology constraints.

Purpose of the Study:

  • To enable visually impaired individuals to better understand natural sciences through tactile graphics of mathematical functions.
  • To develop an advanced model for accurate and efficient recognition and segmentation of function graph elements.
  • To enhance the creation of electronic formats for producing tactile graphics from mathematical functions.

Main Methods:

  • Proposed an MA Mask R-CNN model incorporating MA ConvNeXt for feature extraction and MA BiFPN for feature fusion.
  • Developed novel MA ConvNeXt and MA BiFPN networks to improve feature representation.
  • Integrated an attention mechanism combining local, global, and channel information to enhance detection of diverse targets.

Main Results:

  • The MA Mask R-CNN model achieved 89.6% mAP for target detection and 72.3% mAP for target segmentation in function graph instance segmentation.
  • Demonstrated a 9% improvement in target detection mAP and a 12.8% improvement in target segmentation mAP compared to the original Mask R-CNN.
  • The model effectively enhances detection capabilities for slender and multi-type targets by utilizing multi-scale features.

Conclusions:

  • The MA Mask R-CNN model offers a significant advancement in generating tactile graphics from mathematical functions for the visually impaired.
  • This technology has the potential to improve educational and scientific accessibility for individuals with visual impairments.
  • Further research can explore broader applications of this model in converting complex visual data into accessible formats.