|
Titre :
|
An integrated approach to feature based dynamic vision
|
|
Auteurs :
|
E. Dickmanns ;
IEEE computer vision & pattern recognition conference, Ann Arbor USA, june 5-8, 1988
|
|
Type de document :
|
article/chapitre/communication
|
|
Année de publication :
|
1988
|
|
Format :
|
820-825
|
|
Langues:
|
= Anglais
|
|
Catégories :
|
ROBOTIQUE
|
|
Mots-clés:
|
TRAITEMENT D'IMAGES
;
VISION
|
|
Résumé :
|
A new method for dynamic scene analysis by computer vision is described combining 3D-shape models, dynamical models as known from modern control theory and the laws of perspective projection. In order to arrive at numerically efficient real-time algorithms the recursive state estimation by Kalman filtering is adapted to a feature based image sequence analysis scheme. The spatial and temporal constraint propagation using an integral spatio-temporal model yields image evaluation cycle times of about 0.1 seconds for simple but realistic tasks with microprocessors available today. Motion control in the dynamic range of humans is thereby possible. Applications discussed are : 3 degree of freedom planar docking, road vehicle guidance at speeds up to 60 mph and 6-degree of freedom landing approach of a business jet plane (hardware in the loop simulation).
|
