L’outil de licence Creative Commons du formulaire de soumission ne fonctionne pas actuellement. Pour ajouter une licence CC à votre thèse ou publication, contactez ruor@uottawa.ca après le dépôt. -- The Creative Commons licence tool in the submission form is currently unavailable. To apply a CC licence to your thesis or publication, contact ruor@uottawa.ca after the deposit.

Monocular Obstacle Detection for Moving Vehicles

En cours de chargement...
Vignette d'image

Date

Nom de la revue

ISSN de la revue

Titre du volume

Éditeur

Université d'Ottawa / University of Ottawa

Résumé

This thesis presents a 3D reconstruction approach to the detection of static obstacles from a single rear view parking camera. Corner features are tracked to estimate the vehicle’s motion and to perform multiview triangulation in order to reconstruct the scene. We model the camera motion as planar motion and use the knowledge of the camera pose to efficiently solve motion parameters. Based on the observed motion, we selected snapshots from which the scene is reconstructed. These snapshots guarantee a sufficient baseline between the images and result in more robust scene modeling. Multiview triangulation of a feature is performed only if the feature obeys the epipolar constraint. Triangulated features are semantically labelled according to their 3D location. Obstacle features are spatially clustered to reduce false detections. Finally, the distance to the nearest obstacle cluster is reported to the driver.

Description

Mots-clés

obstacle detection, collision avoidance, 3d reconstruction, triangulation, obstacle detection from moving vehicle

Citation

Approbation

Évaluation

Complété par

Référencé par