Détection de route et suivi de véhicules par vision pour l'ACC

Détection de route et suivi de véhicules par vision pour l'ACC

Road detection and vehicles tracking by vision for ACC

R. Aufrère F. Marmoiton  R. Chapuis  F. Collange  J.P. Dérutin 

Laboratoire des Sciences et Matériaux pour l'Electronique, et d'Automatique (LASMEA) UMR 6602 du CNRS Université Blaise-Pascal de Clermont-Ferrand F-63177 AUBIERE Cedex

Corresponding Author Email: 
aufrere@lasmea.univ-bpclermont.fr
Page: 
233-248
|
Received: 
25 February 2000
| |
Accepted: 
N/A
| | Citation

OPEN ACCESS

Abstract: 

This article deals with a process designed first to extract the lane of vehicle by on-board monocular vision. This detection process is based upon a recursive updating of a statistical model of the lane obtained by a training phase. Once the lane has been located, a reconstruction algorithm computes the vehicle location on its lane and the 3D shape of the road. Thereafter, we are focus at the detection, location and tracking of front vehicles equipped with specific visual markers in order to achieve an accurate determination of the location and speed of these vehicles. Merging these various informations allows to point out the most dangerous obstacle. Each of these three processes is detailed significant examples are provide.

Résumé

Cet article présente, dans un premier temps, un procédé permettant de détecter la voie de circulation d'un véhicule par vision monoculaire embarquée. Ce processus de détection est basé sur une mise à jour récursive d'un modèle statistique de la voie obtenu par une phase d'apprentissage. Après avoir localisé la voie, un algorithme de reconstruction détermine la position du véhicule dans sa voie de circulation et le profil 3D de la route. Par la suite, nous nous intéressons à la détection, la localisation et surtout le suivi des véhicules situés à l'avant et équipés de marques visuelles afin de déterminer avec précision la position et la vitesse relative de ces véhicules. La combinaison de ces différentes informations permet de déterminer le véhicule le plus dangereux. La description détaillée de chacune des étapes de notre algorithme est suivie d'exemples significatifs.

Keywords: 

Road detection, 3D reconstruction, vehicles tracking

Mots clés

Détection de route, reconstruction 3D, suivi de véhicules

1. Introduction
2. Détection Et Reconstruction 3D De La Route
3. Localisation Et Suivi D'obstacles
4. Coopération
5. Implantation
6. Conclusion Et Perspectives
  References

[1] R. Aufrère, R. Chapuis, and F. Chausse. Détection précise d bords de route par vision monocculaire embarquée. In 12` eme congrès Reconnaissance des Formes et Intelligence Artificielle, Paris, France, 1-3 Février 2000.

[2] M. Betke and H. Nguyen. Highway scene analysis from a moving vehicle under reducd visibility conditions. In Processing of the IEEE International Conference o, intelligent Vehicles, volume 1, pages 131–136, Stuttgart, Germany, October 1988.

[3] M. Camiade, D.Domnesque, P.F. Alleaume, A. Mallet, and H. Daemkes. Gaas mmics for automotive radar applications. In Proceeding of thr IEEE Internationam Conference on Intelligent Vehicles, volume 2, pages 371–375, Stuttgart, Germany, October 1998.

[4] R. Chapuis. Suivi de primitives image, application à la conduite automatique sur route. PhD thesis, Univ. B. Pascal, Clermont-Ferrand, France, Janvier 1991.

[5] R. Chapuis, R. Aufrère, and F. Chausse. Recovering a 3d shape of road by vision. In 7th International Conference on Image Processing and its Applications, Manchester (U.K.), July 12-15 1999.

[6] R. Chapuis, A. Potelle, J.L. Brame, and F. Chausse. Real time vehicle trajectory supervision on highway. International Journal of Robotics Research, 14(6) : 531–542, December 1995.

[7] D. DeMenthon. Reconstruction of a road by matching edge points in the road image. In Technical Report CAR-TR-368, Computer Vision Laboratory, Center for Automation Research, University of Maryland, College Park MD20742, June 1998.

[8] D.F. DeMenthon and L.S Davis. New exact and approximate solutions of the three point perspective problem. IEEE Transactions on Pattern analysis and Machine Intelligence, pages 1100–1105, November 1992.

[9] E.D. Dickmanns and V. Graefe. Dynamic monocular machine vision and applications of dynamic monocular vision. International Journal of Machine Vision and Application,1 : 223–240 and 241–261, 1988.

[10] K. Kanatani and K. Watanabe. Reconstruction of 3d road geometry from images for autonomous land vehicles. IEEE Trans. On Robotics and Automation, 6 : 127–132, February 1990.

[11] C Kreucher, S. Lakshmanan, and K. Kluge. A driver warning system based on th lois lane detection algorithm. In Proceeding of the IEEE International Conference on Intelligent Vehicles, volume 1, pages 17–22, Stuttgart, Germany, October 1998.

[12] F. Marmoiton, F. Collange, J. Alizon, and J.P. Dérutin. 3d localization of a car observed through a monocular video camera. In Proceedings of the IEEE International Conference on Intelligent Vehicles, volume 1, pages 189–194, Stuttgart, Germany, October 1998..

[13] N.D. Mattews, P.E.An, D. Charnley, and C.J. Harris. Vehicle detection and recognition in greyscale imagery. In 2d symposium on Intelligent Autonomous Vehicle, pages 31–36, Espoo, Finland, June 1995.

[14] E. Montagne, J. Alizon, P. Martinet, and J. Gallice. Real time 3D location of a car from three characteristic points observed in a video image sequence. In 7th Symposium on Transportation Systems, volume 2, pages 385–390, Tianjin, China, August.

[15] S. Schultz. A modeled based codesign applications : the design of an autonomous intelligent cruise controller. PhD thesis, Department of Electrical and Computer Engineering, University of Arizona, USA. 1997.

[16] L.T. Shaaser and B.T Thomas. Finding road lane boundaries for vision guided navigation. Proceedings of the Rountable discussion on Vision Vehicle Guidance, pages 10.1–10.9, 1990.

[17] Y. Bar Shalom and T.E. Fortman. Tracking and data association, volume 179. Academic Press, INC., 1988.

[18] L. Trassoudaine, J. Alizon, J. Gallice, and F. Collange. Visual tracking by a multisensorial approach. In 1st Workshop on Intelligent Autonomous Vehicles, pages 113–118, Southampton, April 1993.

[19] T. Zielke, M. Braukmann, and W. Von Steelen. Intensity and edge based symmetry detction with an application to car following. In Computer Vision Graphics and Image Processing, Image Understanding, volume 58(2), pages 177–190, September 1993.