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

An Inverse Problem for a Fractional Space-Time Diffusion Equation with Fractional Boundary Condition.

Entropy (Basel, Switzerland)·2026
Same author

Prevalence and morphometric characteristics of the psoas minor muscle in human fetuses: an anatomical study.

Folia morphologica·2025
Same author

Remote Eye Movement Desensitization and Reprocessing Treatment of Long-COVID- and Post-COVID-Related Traumatic Disorders: An Innovative Approach.

Brain sciences·2025
Same author

Inferring About the Average Value of Audit Errors from Sequential Ratio Tests.

Entropy (Basel, Switzerland)·2024
Same author

Convolutional Neural Network Model for Intestinal Metaplasia Recognition in Gastric Corpus Using Endoscopic Image Patches.

Diagnostics (Basel, Switzerland)·2024
Same author

Design of Laboratory Stand for Displacement Measurement of IPMC Actuators.

Sensors (Basel, Switzerland)·2023

Related Experiment Video

Updated: Jan 7, 2026

Author Spotlight: Revolutionizing Remote Surgery with Augmented Reality and Robotics for Enhanced Precision and Accessibility
07:46

Author Spotlight: Revolutionizing Remote Surgery with Augmented Reality and Robotics for Enhanced Precision and Accessibility

Published on: August 9, 2024

1.2K

Handwritten Digit Recognition with Flood Simulation and Topological Feature Extraction.

Rafał Brociek1, Mariusz Pleszczyński2, Jakub Błaszczyk3

  • 1Department of Artificial Intelligence Modelling, Faculty of Applied Mathematics, Silesian University of Technology, 44-100 Gliwice, Poland.

Entropy (Basel, Switzerland)
|December 24, 2025
PubMed
Summary

This study presents a new handwritten digit recognition method using flood simulation and topology. It accurately classifies digits, outperforming traditional methods with less data and noise.

Keywords:
classificationcomputer visionimage recognitionneural networkspattern recognitionvector databasewater simulation

More Related Videos

Microfluidic Platform with Multiplexed Electronic Detection for Spatial Tracking of Particles
11:54

Microfluidic Platform with Multiplexed Electronic Detection for Spatial Tracking of Particles

Published on: March 13, 2017

9.7K
Capturing Dynamic Finger Gesturing with High-resolution Surface Electromyography and Computer Vision
08:15

Capturing Dynamic Finger Gesturing with High-resolution Surface Electromyography and Computer Vision

Published on: March 28, 2025

1.1K

Related Experiment Videos

Last Updated: Jan 7, 2026

Author Spotlight: Revolutionizing Remote Surgery with Augmented Reality and Robotics for Enhanced Precision and Accessibility
07:46

Author Spotlight: Revolutionizing Remote Surgery with Augmented Reality and Robotics for Enhanced Precision and Accessibility

Published on: August 9, 2024

1.2K
Microfluidic Platform with Multiplexed Electronic Detection for Spatial Tracking of Particles
11:54

Microfluidic Platform with Multiplexed Electronic Detection for Spatial Tracking of Particles

Published on: March 13, 2017

9.7K
Capturing Dynamic Finger Gesturing with High-resolution Surface Electromyography and Computer Vision
08:15

Capturing Dynamic Finger Gesturing with High-resolution Surface Electromyography and Computer Vision

Published on: March 28, 2025

1.1K

Area of Science:

  • Computer Vision
  • Machine Learning
  • Pattern Recognition

Background:

  • Traditional pixel-based digit recognition methods struggle with noise and limited data.
  • Existing approaches often lack robustness to variations like rotation and occlusion.

Purpose of the Study:

  • Introduce a novel topology-driven approach for handwritten digit recognition.
  • Enhance classification accuracy and robustness against common image imperfections.

Main Methods:

  • Simulate directional flood from image boundaries using a modified breadth-first search (BFS) algorithm.
  • Extract topological features including stroke directionality, segmentation, and closed areas.
  • Utilize Annoy approximate nearest neighbors for efficient classification.

Main Results:

  • Achieved 95.9% accuracy on the MNIST dataset and 93.0% on the USPS dataset.
  • Demonstrated resilience to rotation, noise, and limited training data.
  • Reduced feature dimensionality while improving generalization.

Conclusions:

  • The proposed topology-driven digit recognition method offers high accuracy and robustness.
  • This approach provides a compact and interpretable feature representation for classification.
  • Flood simulation and topological feature extraction present a promising alternative to pixel-based methods.