Real-Time Adaptive Multi-Classifier Multi-Resolution Visual Tracking Framework for Unmanned Aerial Vehicles

Fu, Changhong; Suárez Fernández, Ramón; Olivares Méndez, Miguel Ángel y Campoy Cervera, Pascual (2013). Real-Time Adaptive Multi-Classifier Multi-Resolution Visual Tracking Framework for Unmanned Aerial Vehicles. En: "2nd IFAC Workshop on Research, Education and Development of Unmanned Aerial Systems", 20-22 Nov 2013, Compiegne, France. ISBN 978-3-902823-57-1. pp. 99-106.

Descripción

Título: Real-Time Adaptive Multi-Classifier Multi-Resolution Visual Tracking Framework for Unmanned Aerial Vehicles
Autor/es:
  • Fu, Changhong
  • Suárez Fernández, Ramón
  • Olivares Méndez, Miguel Ángel
  • Campoy Cervera, Pascual
Tipo de Documento: Ponencia en Congreso o Jornada (Artículo)
Título del Evento: 2nd IFAC Workshop on Research, Education and Development of Unmanned Aerial Systems
Fechas del Evento: 20-22 Nov 2013
Lugar del Evento: Compiegne, France
Título del Libro: 2nd IFAC Workshop on Research, Education and Development of Unmanned Aerial Systems
Fecha: 2013
ISBN: 978-3-902823-57-1
Materias:
Palabras Clave Informales: Unmanned Aerial Vehicles(UAVs), Discriminative Visual Tracking(DVT), Hierarchical Tracking Strategy(HTS), Online Appearance Learning(OAL), Compressive Visual Sensing(CVS), Adaptive Algorithm, Robot Navigation.
Escuela: E.T.S.I. Industriales (UPM)
Departamento: Automática, Ingeniería Electrónica e Informática Industrial [hasta 2014]
Grupo Investigación UPM: Computer Vision Group
Licencias Creative Commons: Ninguna

Texto completo

[img]
Vista Previa
PDF (Document Portable Format) - Se necesita un visor de ficheros PDF, como GSview, Xpdf o Adobe Acrobat Reader
Descargar (1MB) | Vista Previa

Resumen

This paper presents a novel robust visual tracking framework, based on discriminative method, for Unmanned Aerial Vehicles (UAVs) to track an arbitrary 2D/3D target at real-time frame rates, that is called the Adaptive Multi-Classifier Multi-Resolution (AMCMR) framework. In this framework, adaptive Multiple Classifiers (MC) are updated in the (k-1)th frame-based Multiple Resolutions (MR) structure with compressed positive and negative samples, and then applied them in the kth frame-based Multiple Resolutions (MR) structure to detect the current target. The sample importance has been integrated into this framework to improve the tracking stability and accuracy. The performance of this framework was evaluated with the Ground Truth (GT) in different types of public image databases and real flight-based aerial image datasets firstly, then the framework has been applied in the UAV to inspect the Offshore Floating Platform (OFP). The evaluation and application results show that this framework is more robust, efficient and accurate against the existing state-of-art trackers, overcoming the problems generated by the challenging situations such as obvious appearance change, variant illumination, partial/full target occlusion, blur motion, rapid pose variation and onboard mechanical vibration, among others. To our best knowledge, this is the first work to present this framework for solving the online learning and tracking freewill 2D/3D target problems, and applied it in the UAVs.

Proyectos asociados

TipoCódigoAcrónimoResponsableTítulo
Gobierno de EspañaPI2010-20751-C02-01CICYTSin especificarSin especificar
FP7FP7-ICT-231143ECHORDSin especificarOMNIWORKS project

Más información

ID de Registro: 37650
Identificador DC: http://oa.upm.es/37650/
Identificador OAI: oai:oa.upm.es:37650
URL Oficial: http://www.ifac-papersonline.net/
Depositado por: Memoria Investigacion
Depositado el: 14 Sep 2015 15:40
Ultima Modificación: 16 Sep 2015 15:42
  • Open Access
  • Open Access
  • Sherpa-Romeo
    Compruebe si la revista anglosajona en la que ha publicado un artículo permite también su publicación en abierto.
  • Dulcinea
    Compruebe si la revista española en la que ha publicado un artículo permite también su publicación en abierto.
  • Recolecta
  • e-ciencia
  • Observatorio I+D+i UPM
  • OpenCourseWare UPM