Skip to Main content Skip to Navigation

Système de vision hybride : modélisation et application au suivi haute résolution

Abstract : This thesis deals with the automatic tracking person through the implementation of a hybrid system combining a vision sensor with a large field of view and low resolution device with a Pan-Tilt-Zoom adjustable ensuring the detail of the information . The first part is devoted to the original proposal of a solution when calibrating the optical centers of the two sensors are practically alike. In a second part, we discuss the linking of information from each sensor and mutual enrichment needed for monitoring person relationship based on a particle filtering approach: the prediction step of the filter is guided by the detection result the static camera and the measurement is given by a linear appearance of the target derived from the dynamic camera. Finally, the first monitoring results are presented and analyzed.
Document type :
Complete list of metadata

Cited literature [114 references]  Display  Hide  Download
Contributor : Camille Meyer Connect in order to contact the contributor
Submitted on : Tuesday, September 11, 2012 - 4:13:42 PM
Last modification on : Wednesday, April 21, 2021 - 8:34:02 AM
Long-term archiving on: : Friday, December 16, 2016 - 12:10:40 PM


  • HAL Id : tel-00730993, version 1


Julie Badri. Système de vision hybride : modélisation et application au suivi haute résolution. Automatique / Robotique. Université Blaise Pascal - Clermont-Ferrand II, 2008. Français. ⟨NNT : 2008CLF21866⟩. ⟨tel-00730993⟩



Record views


Files downloads