Skip to Main content Skip to Navigation

A STATISTICAL FRAMEWORK FOR POSITIVE DATA CLUSTERING WITH FEATURE SELECTION : APPLICATION TO OBJECT DETECTION

Abstract : In this paper, we concern ourselves with the problem of simultaneous positive data clustering and feature selection. We propose a statistical framework based on finite mixture models of generalized inverted Dirichlet (GID) distributions. The GID offers a more practical and flexible alternative to the inverted Dirichlet which has a very restrictive covariance structure. For learning the parameters of the resulting mixture, we propose an approach based on minimum message length (MML) criterion. We use synthetic data and real data generated from a challenging application that concerns objects detection to demonstrate the feasibility and advantages of the proposed method.
Complete list of metadatas

Cited literature [13 references]  Display  Hide  Download

https://hal.inria.fr/hal-00908379
Contributor : Khalid Daoudi <>
Submitted on : Friday, November 22, 2013 - 5:26:25 PM
Last modification on : Monday, January 13, 2020 - 10:04:03 AM
Document(s) archivé(s) le : Sunday, February 23, 2014 - 4:31:20 AM

File

Eusipco2013.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00908379, version 1

Collections

Citation

Mohamed Al Mashrgy, Nizar Bouguila, Khalid Daoudi. A STATISTICAL FRAMEWORK FOR POSITIVE DATA CLUSTERING WITH FEATURE SELECTION : APPLICATION TO OBJECT DETECTION. Eusipco, Sep 2013, Marrakech, Morocco. ⟨hal-00908379⟩

Share

Metrics

Record views

917

Files downloads

576