yes - Site CSC en Construction

Document technical information

Format pdf
Size 323.3 kB
First found May 22, 2018

Document content analysis

Category Also themed
Language
English
Type
not defined
Concepts
no text concepts found

Persons

Organizations

Places

Transcript

PhD Grants from the China Scholarship Council
Co-operation Program with the UT and INSA
Program 2017
Vehicle localization based on scene perception
UTBM – IRTES/SeT (ICAP Team – PENA Group)
http://epan.fr/Accueil
PhD subject description:
The PhD proposition is a part of the "Intelligent Vehicle" project of the IRTES-SET laboratory that
aims at proposing advanced concepts for the development of autonomous vehicles and driver
assistance systems. To achieve intelligent or autonomous navigation, several functions are required,
among them vehicle ego-localization and environment perception and analysis. In this framework,
our team develops research activities especially in image processing, computer vision, multi-sensor
based localization and perception, artificial intelligence ...
In urban environments, the use of exclusively GPS is not sufficient to achieve robust and accurate
localization. In this thesis, we propose then to take advantage of perception sensors (as camera,
laser range finder) for vehicle localization. The principle of the envisaged approach is to compare
the current perception of the scene with a geo-referenced database. This database could be obtained:
- from a learning phase, during which the sensors data are acquired and stored. In that case, a step of
vision based 3D reconstruction of the scene can be envisaged (this reconstructed 3D model could
even be used to improve GPS accuracy using correction models).
- or from an open source as Google Street View.
The objective of this comparison is to find the reference image or data that corresponds to the
current scene view.
Keywords:
Vehicle localization, google street view, image retrieval, machine learning, 3D reconstruction
Candidate profile:
The candidate should have skills in the following disciplines: signal processing, applied
mathematics, computer vision/image processing, machine learning. Strong skills in Matlab and
C/C++ programming are also required. The candidate will be involved within an experimental
platform (intelligent automated vehicles equipped with different sensors) and should participate to
acquire data, test and evaluate developed algorithms/methods. For a description of the vehicles:
http://epan.fr/Véhicule_intelligent
Experimental platform:
The PhD student will benefit from the SeT experimental platform of electric and automated
vehicles, equipped with cameras, stereoscopic systems, laser range finders, LIDARs, GPS, inertial
navigation systems, gyrometers, …
Selected papers recently published by the team (2014-2015):
2015

Combination of Block-Based Image Matching Using Local Binary Patterns and LIDAR Data
for Vehicle Localization
Yongliang QIAO, Cindy CAPPELLE, Yassine RUICHEK, and Fadi DORNAIKA.
In Proc. of 3rd CEAS EuroGNC Conference, Toulouse, France, 2015.

Multiframe-Based High Dynamic Range Monocular Vision System for Advanced Driver
Assistance Systems
You LI, Yongliang QIAO, and Yassine RUICHEK.
In IEEE Sensors Journal, vol. 15(10), pp. 5433 - 5441, IEEE, 2015.

Place Recognition Based Visual Localization Using LBP Feature and SVM
Yongliang QIAO, Cindy CAPPELLE, and Yassine RUICHEK.
In Proc. of 14th Mexican International Conference on Artificial Intelligence (MICAI'2015), Lecture Notes in
Artificial Intelligence (LNAI), Springer, vol. 9414, Cuernavaca, Mexico, 2015.

Video and LIDAR Based Object Tracking Using Small-Region Growth
Yong FANG, Cindy CAPPELLE, and Yassine RUICHEK.
In Proc. of IEEE International Conference on Intelligent Transportation Systems (ITSC'2015), pp. 2323-2328,
Canary Islands, Spain, 2015.
2014

3D Modeling of Urban Environments for Enhancing GPS Localization's Accuracy
Julien MOREAU, Sébastien AMBELLOUIS, and Yassine RUICHEK.
In Proc. of Transport Research Arena (TRA'2014), Paris, France, 2014.

All-Day Moving Objects Detection for Security at Level Crossing
Julian MURGIA, Cyril MEURIE, and Yassine RUICHEK.
In Proc. of Transport Research Arena (TRA'2014), Paris, France, 2014.

An Improved Colorimetric Invariants and Rgb-Depth-Based Codebook Model for Background
Subtraction Using Kinect
Julian MURGIA, Cyril MEURIE, and Yassine RUICHEK.
In Proc. of 13th Mexican International Conference on Artificial Intelligence (MICAI'2014), Lecture Notes in
Artificial Intelligence (LNAI), Springer, vol. 8856, pp. 380-392, Tuxtla Gutiérrez, Chiapas, Mexico, 2014.

A Novel Evidence Based Model for Detecting Dangerous Situations in Level Crossing
Environments
Houssam SALMANE, Yassine RUICHEK, and Louahdi KHOUDOUR.
In Journal of Expert Systems With Applications, vol. 41(3), pp. 795-810, 2014.

Extrinsic Calibration Between 2D Laser Range Finder and Fisheye Camera
Yong FANG, Cindy CAPPELLE, and Yassine RUICHEK.
In Proc. of 10th International Symposium on Visual Computing (ISVC'2014), Lecture Notes in Computer
Science (LNCS), Springer, vol. 8888, pp. 925-935, Las Vegas, Nevada, USA, 2014.

Gnss Reflection Mitigation by 3d Urban Structure Estimation
Juliette MARAIS, Julien MOREAU, Sébastien AMBELLOUIS, and Yassine RUICHEK.
Workshop Hybridization GNSS and Video, ENAC, Toulouse, 2014.

Image Based Place Recognition and Lidar Validation for Vehicle Localization
Yongliang QIAO, Cindy CAPPELLE, and Yassine RUICHEK.
In Proc. of 13th Mexican International Conference on Artificial Intelligence (MICAI'2014), Lecture Notes in
Artificial Intelligence (LNAI), Springer, vol. 8856, pp. 304-315, Tuxtla Gutiérrez, Chiapas, Mexico, 2014.

Locality Constrained Encoding Graph Construction and Application to Outdoor Object
Classification
Fadi DORNAIKA, Alireza BOSAGHZADEH, Houssam SALMANE, and Yassine RUICHEK.
In Proc. of 22nd International Conference on Pattern Recognition (ICPR'2014), Stockholm, Sweden, 2014.

Multisensor Based Obstacle Detection in Challenging Scenes
Yong FANG, Cindy CAPPELLE, and Yassine RUICHEK.
In Proc. of 13th Mexican International Conference on Artificial Intelligence (MICAI'2014), Lecture Notes in
Artificial Intelligence (LNAI), Springer, vol. 8856, pp. 257-268, Tuxtla Gutiérrez, Chiapas, Mexico, 2014.

Perception of Dynamic Urban Environments Using Stereovision Based Dynamic Occupancy
Grid Map
You LI, Yassine RUICHEK.
In Proc. of Transport Research Arena (TRA'2014), Paris, France, 2014.

Road Detection Using Fisheye Camera and Laser Range Finder
Yong FANG, Cindy CAPPELLE, and Yassine RUICHEK.
In Proc. of IAPR International Conference on Image and Signal Processing (ICISP'2014), Lecture Notes in
Computer Science (LNCS), Springer, vol. 8509, pp. 495-502, Cherbourg, France, 2014.

Camera/laser/gps Fusion Method for Vehicle Positioning Under Extended Nis Based Sensor
Validation
Lijun WEI, Cindy CAPPELLE, and Yassine RUICHEK.
In IEEE Transactions on Instrumentation & Measurement, vol. 62(11), pp. 3110-3122, 2013.
×

Report this document