Perception monoculaire de l'environnement pour les systèmes de transport intelligents

Abstract : The evolution of transport in recent decades, demonstrates an ongoing commitment to reduce the constraints associated with the concept of displacement. To this end, an important part of efforts intended to reduce the travel time, mainly through the improvement of infrastructure and alternative transportation. The multi-modal transportation planning, expected to meet different needs of users, was not enough to stop the rise of the automobile in urban areas. Thus, cars are gradually become the main source of pollution and urban accidents. The solutions studied to remedy this situation are mainly based on the responsibility of the human factor. They essentially offer to replace the automobile by autonomous transport system. The automation of vehicles, gradually established by the Advanced Driving Assistance Systems, requires the development of modules for environmental perception that analyze and process information gained from one or more sensors. With the boom of computational capabilities embedded systems, the camera has become one of the most used sensors, as long for the wealth of information contained in a sequence of images, for its low cost and small footprint. The work presented in this thesis provides a novel solution to the problem of visual perception for automatic driving through a monocular approach based on the study of geometric constraints applied to motion picture.
Document type :
Theses
Domain :
Complete list of metadatas

Cited literature [98 references]  Display  Hide  Download

https://pastel.archives-ouvertes.fr/pastel-00005607
Contributor : Ecole Mines Paristech <>
Submitted on : Wednesday, December 2, 2009 - 8:00:00 AM
Last modification on : Monday, November 12, 2018 - 10:57:00 AM
Long-term archiving on: Friday, September 10, 2010 - 2:46:06 PM

Identifiers

  • HAL Id : pastel-00005607, version 1

Citation

Yann Dumortier. Perception monoculaire de l'environnement pour les systèmes de transport intelligents. domain_other. École Nationale Supérieure des Mines de Paris, 2009. Français. ⟨NNT : 2009ENMP1640⟩. ⟨pastel-00005607⟩

Share

Metrics

Record views

3716

Files downloads

1930