Online multi-model particle filter-based tracking to study bedload transport

Hugo Lafaye de Micheaux 1, 2 Christophe Ducottet 1 Philippe Frey 2
1 LHC - Laboratoire Hubert Curien [Saint Etienne]
2 UR ETGR - Erosion torrentielle neige et avalanches
Abstract : Multi-object tracking is a difficult problem underlying many computer vision applications. In this work, we focus on sediment transport experiments in a flow were sediments are represented by spherical calibrated beads. The aim is to track all beads over long time sequences to obtain sediment velocities and concentration. Classical algorithms used in fluid mechanics fail to track the beads over long sequences with a high precision because they incorrectly handle both miss-detections and detector imprecision. Our contribution is to propose a particle filter-based algorithm including an adapted multiple motion model. Additionally, this algorithm integrates several improvements to account for the lack of precision of the detector. The evaluation was made using a test sequence with a dedicated ground-truth. The results show that the method outperforms state-of-the-art concurrent algorithms.
Type de document :
Poster
IEEE International Conference on Image Processing (ICIP 2016), Sep 2016, Phoenix, AZ, United States. 2016
Liste complète des métadonnées
https://hal-ujm.archives-ouvertes.fr/ujm-01491863
Contributeur : Hugo Lafaye de Micheaux <>
Soumis le : vendredi 17 mars 2017 - 14:52:23
Dernière modification le : mardi 1 mai 2018 - 11:13:58
Document(s) archivé(s) le : dimanche 18 juin 2017 - 13:22:18

Fichier

poster_icip_2016_lafaye.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : ujm-01491863, version 1

Collections

UNIV-ST-ETIENNE | IRSTEA

Relations

Citation

Hugo Lafaye de Micheaux, Christophe Ducottet, Philippe Frey. Online multi-model particle filter-based tracking to study bedload transport. IEEE International Conference on Image Processing (ICIP 2016), Sep 2016, Phoenix, AZ, United States. 2016. 〈ujm-01491863〉

Exporter

BibTeX TEI DC DCterms EndNote

Partager

Métriques

Consultations de la notice

57

Téléchargements de fichiers

37

Contact

support.ccsd.cnrs.fr
hal.support@ccsd.cnrs.fr
Logo CCSD