EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Enhancing symbolic image classification through Gaussian copulas and optimized distinguishing points

Sri Winarni iD, Sapto Wahyu Indratno, Mohd Shahizan Othman iD, Siti Zaiton Mohd Hashim, Mohd Murtadha Mohamad, Apri Junaidi, Ebenezer Bonyah, Anindya Apriliyanti Pravitasari, Triyani Hendrawati, Irlandia Ginanjar

PLOS One · 20.04.2026 · 21 (4) · e0346790

DOI: 10.1371/journal.pone.0346790 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS One, online erschienen am 20.04.2026. Verfasst von Sri Winarni und 9 weiteren Autor:innen. Beteiligte Einrichtung: Department of Statistics, Universitas Padjadjaran, Sumedang, Indonesia. Hauptthemen laut MEDLINE: Image Processing, Computer-Assisted, Pattern Recognition, Automated. Die Arbeit nennt 51 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Universitas Padjadjaran.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In this paper a new method is being proposed which makes use of symbolic data manifested by empirical cumulative distribution functions (ECDF) and distribution functions based on sets of ECDF values, referred to as the distribution function of distribution values (DFDV) of image features. This differs with conventional image classification studies, which mostly rely on pixel intensity values as features. Such symbolic representations provide a more general characterization of the pixel intensities patterns across image areas. The main novelty associated with the given research is the creation of the DFDV based on the best possible selection of the distinguishing points in order to capture fundamental changes in intensity distributions within image classes. Compared to previous studies which have used pre-determined distinguishing points, the proposed study proposes a clustering-based approach to find distinguishing points that maximise class separability. Experimental evaluation on the MNIST handwritten digits data set demonstrates the effectiveness of the proposed method, achieving an average classification accuracy of 68.27% and a highest accuracy of 95.33%. These results indicate that integrating clustering-based symbolic features extraction with copula-based modelling provides a competitive and promising for image classification tasks.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Image Processing, Computer-AssistedPattern Recognition, AutomatedClustering AlgorithmsNormal DistributionClassification AlgorithmsCluster AnalysisAlgorithms

Autor:innen und Institutionen

  • Department of Statistics, Universitas Padjadjaran, Sumedang, Indonesia.
  • Statistics Research Group, Institut Teknologi Bandung, Bandung, Indonesia.
  • Faculty of Computing, Universiti Teknologi Malaysia, Johor Bahru, Malaysia.
  • Department of Mathematics Education, Akenten Appiah-Menka University of Skills Training and Entrepreneurial Development, Kumasi, Ghana.
  • Azerbaijan University of Architecture and Construction, Baku, Azerbaijan.

ORCID-Profile: Sri Winarni, Mohd Shahizan Othman

Förderung

  • Universitas Padjadjaran

Zitieren

APA

Winarni, S., Indratno, S. W., Othman, M. S., Hashim, S. Z. M., Mohamad, M. M., Junaidi, A., Bonyah, E., Pravitasari, A. A., Hendrawati, T., & Ginanjar, I. (2026). Enhancing symbolic image classification through Gaussian copulas and optimized distinguishing points. PLOS One, 21(4), e0346790. https://doi.org/10.1371/journal.pone.0346790

Vancouver

Winarni S, Indratno SW, Othman MS, Hashim SZM, Mohamad MM, Junaidi A, et al. Enhancing symbolic image classification through Gaussian copulas and optimized distinguishing points. PLOS One. 2026;21(4):e0346790. doi:10.1371/journal.pone.0346790

BibTeX
@article{Winarni2026,
  author = {Winarni, Sri and Indratno, Sapto Wahyu and Othman, Mohd Shahizan and Hashim, Siti Zaiton Mohd and Mohamad, Mohd Murtadha and Junaidi, Apri and Bonyah, Ebenezer and Pravitasari, Anindya Apriliyanti and Hendrawati, Triyani and Ginanjar, Irlandia},
  title = {Enhancing symbolic image classification through Gaussian copulas and optimized distinguishing points},
  journal = {PLOS One},
  year = {2026},
  volume = {21},
  number = {4},
  pages = {e0346790},
  doi = {10.1371/journal.pone.0346790}
}
Kostenlos starten →