EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Handwritten Digit Recognition with Flood Simulation and Topological Feature Extraction

Rafał Brociek iD, Mariusz Pleszczyński iD, Jakub Błaszczyk, Maciej Czaicki iD, Christian Napoli iD

Entropy · 29.11.2025 · 27 (12) · 1218

DOI: 10.3390/e27121218 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Entropy, online erschienen am 29.11.2025. Verfasst von Rafał Brociek und 4 weiteren Autor:innen. Beteiligte Einrichtung: Department of Artificial Intelligence Modelling, Faculty of Applied Mathematics, Silesian University of Technology, 44-100 Gliwice, Poland. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This paper introduces a novel approach to handwritten digit recognition based on directional flood simulation and topological feature extraction. While traditional pixel-based methods often struggle with noise, partial occlusion, and limited data, our method leverages the structural integrity of digits by simulating water flow from image boundaries using a modified breadth-first search (BFS) algorithm. The resulting flooded regions capture stroke directionality, spatial segmentation, and closed-area characteristics, forming a compact and interpretable feature vector. Additional parameters such as inner cavities, perimeter estimation, and normalized stroke density enhance classification robustness. For efficient prediction, we employ the Annoy approximate nearest neighbors algorithm using ensemble-based tree partitioning. The proposed method achieves high accuracy on the MNIST (95.9%) and USPS (93.0%) datasets, demonstrating resilience to rotation, noise, and limited training data. This topology-driven strategy enables accurate digit classification with reduced dimensionality and improved generalization.

Schlagwörter

Stichwörter der Autor:innen

classificationcomputer visionimage recognitionneural networkspattern recognitionvector databasewater simulation

Autor:innen und Institutionen

  • Department of Artificial Intelligence Modelling, Faculty of Applied Mathematics, Silesian University of Technology, 44-100 Gliwice, Poland.
  • Department of Mathematical Methods in Technology and Computer Science, Faculty of Applied Mathematics, Silesian University of Technology, 44-100 Gliwice, Poland.
  • Faculty of Applied Mathematics, Silesian University of Technology, 44-100 Gliwice, Poland.
  • Department of Computer, Control, and Management Engineering, Sapienza University of Rome, Via Ariosto 25, 00185 Roma, Italy.
  • Department of Artificial Intelligence, Czestochowa University of Technology, Dabrowskiego 69, 42-201 Czestochowa, Poland.

ORCID-Profile: Rafał Brociek, Mariusz Pleszczyński, Maciej Czaicki, Christian Napoli

Zitieren

APA

Brociek, R., Pleszczyński, M., Błaszczyk, J., Czaicki, M., & Napoli, C. (2025). Handwritten Digit Recognition with Flood Simulation and Topological Feature Extraction. Entropy, 27(12), 1218. https://doi.org/10.3390/e27121218

Vancouver

Brociek R, Pleszczyński M, Błaszczyk J, Czaicki M, Napoli C. Handwritten Digit Recognition with Flood Simulation and Topological Feature Extraction. Entropy. 2025;27(12):1218. doi:10.3390/e27121218

BibTeX
@article{Brociek2025,
  author = {Brociek, Rafał and Pleszczyński, Mariusz and Błaszczyk, Jakub and Czaicki, Maciej and Napoli, Christian},
  title = {Handwritten Digit Recognition with Flood Simulation and Topological Feature Extraction},
  journal = {Entropy},
  year = {2025},
  volume = {27},
  number = {12},
  pages = {1218},
  doi = {10.3390/e27121218}
}
Kostenlos starten →