Skip to main content
main-content

Tipp

Weitere Artikel dieser Ausgabe durch Wischen aufrufen

14.02.2020 | Methodologies and Application | Ausgabe 17/2020

Soft Computing 17/2020

Multivariate bounded support Laplace mixture model

Zeitschrift:
Soft Computing > Ausgabe 17/2020
Autoren:
Muhammad Azam, Nizar Bouguila
Wichtige Hinweise
Communicated by V. Loia.

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Abstract

In this paper, bounded Laplace mixture model (BLMM) is proposed. The parameters of proposed model are estimated by maximum likelihood approach via expectation maximization and Newton–Raphson algorithm. The model is proposed for data modeling to perform clustering using synthetic data for univariate and multivariate examples and real datasets of different medical experiments. BLMM is validated through correctness of estimated parameters for synthetic data and clustering accuracy of medical datasets. A new modeling scheme is also introduced for wavelet coefficients which is based on BLMM. It is applied to image clustering and content-based image retrieval (CBIR) for feature extraction in wavelet domain. For feature extraction in this application, each image is decomposed into a set of wavelet subspaces and BLMM with two components is adopted to model the statistical characteristics of the wavelet coefficients for each wavelet subspace. The model parameters adapted from BLMM represent the image features in wavelet domain for each subspace and selected to formulate the feature space which is further used in clustering and CBIR. In the framework for clustering and image retrieval, features extracted in wavelet domain are further modeled through BLMM to categorize images into different groups and trained model is adopted for CBIR. In order to perform image retrieval with trained model via BLMM, city block distance, posterior probability and Kullback–Leibler divergence are introduced. We also propose a novel solution to compute Kullback–Leibler divergence which is very effective for image retrieval due to its low computational complexity and high retrieval rate. The effectiveness and viability of BLMM in texture image clustering and CBIR are demonstrated through UIUC, KTH-TIPS, DTD, STex and Kylberg databases. Different experiments are performed in the chosen applications, and from the results, BLMM has demonstrated its effectiveness in modeling synthetic data, real datasets from medical experiments, feature extraction in wavelet domain, image clustering and CBIR.

Bitte loggen Sie sich ein, um Zugang zu diesem Inhalt zu erhalten

Sie möchten Zugang zu diesem Inhalt erhalten? Dann informieren Sie sich jetzt über unsere Produkte:

Springer Professional "Wirtschaft+Technik"

Online-Abonnement

Mit Springer Professional "Wirtschaft+Technik" erhalten Sie Zugriff auf:

  • über 69.000 Bücher
  • über 500 Zeitschriften

aus folgenden Fachgebieten:

  • Automobil + Motoren
  • Bauwesen + Immobilien
  • Business IT + Informatik
  • Elektrotechnik + Elektronik
  • Energie + Umwelt
  • Finance + Banking
  • Management + Führung
  • Marketing + Vertrieb
  • Maschinenbau + Werkstoffe
  • Versicherung + Risiko

Testen Sie jetzt 30 Tage kostenlos.

Springer Professional "Wirtschaft"

Online-Abonnement

Mit Springer Professional "Wirtschaft" erhalten Sie Zugriff auf:

  • über 58.000 Bücher
  • über 300 Zeitschriften

aus folgenden Fachgebieten:

  • Bauwesen + Immobilien
  • Business IT + Informatik
  • Finance + Banking
  • Management + Führung
  • Marketing + Vertrieb
  • Versicherung + Risiko




Testen Sie jetzt 30 Tage kostenlos.

Springer Professional "Technik"

Online-Abonnement

Mit Springer Professional "Technik" erhalten Sie Zugriff auf:

  • über 50.000 Bücher
  • über 380 Zeitschriften

aus folgenden Fachgebieten:

  • Automobil + Motoren
  • Bauwesen + Immobilien
  • Business IT + Informatik
  • Elektrotechnik + Elektronik
  • Energie + Umwelt
  • Maschinenbau + Werkstoffe




Testen Sie jetzt 30 Tage kostenlos.

Literatur
Über diesen Artikel

Weitere Artikel der Ausgabe 17/2020

Soft Computing 17/2020 Zur Ausgabe

Premium Partner

    Bildnachweise