Layachi Bentabet (Ph.D.)

Tel. 819-822-9600. ext. 2441
Fax. 819-822-9661
e-mail: lbentabe@ubishops.ca.

Profile
Research interests
Publications
Teaching

Profile

I am an assistant professor in the Department of Computer Science at Bishop's University. I received my Engineer degree in Electronic from the National Polytechnic School in Algeria, my Master’s degree in Signal, Image, and Speech Analysis from Institut National des Sciences Appliquées de Lyon in France, and my Ph.D in Computer Science from Université de Sherbrooke in Canada .I am a member of MOIVRE group (Modelisation en Imagerie, Vision et Reseaux de neurones) and my research activities are in the fields of image processing and computer vision

Download my CV (PDF )

Up

Research interests

•    Pattern Recognition and Remote Sensing
•    Data Fusion Techniques for Decision Making
•    PDE-Level Set, Lagrangian and Variational Methods
•    Multivariate Statistical Analysis and Energy-minimization Approaches
•    Application of Topological Methods in Image Processing and Computer Vision

Current Projects

Currently I am supervising and co-supervising graduate students that are working and contributing in the following topics

Demspter-Shafer theory: we are developing automatic methods of determining mass functions in Dempster-Shafer theory from fuzzy memberships functions and probabilities. Our approaches are used in tackling problems arising in medical imaging such as finding defects and anomalies in certain organs and in remote sensing for fusion of radar and optical images.

A radar image

An optical image

The fusion result (with 6 clusters)

Java Applet (coming soon)

Spatiotemporal modeling using multiple cameras: in this topic, we are concerned with the spatiotemporal modeling of the environment (segmentation, detection, and tracking) using multiple cameras. We are especially interested in modeling the sensors fusion and decision making mechanisms in situations where sensors produce incomplete and ambiguous description of the scene. These situations are usually caused by high occlusion, changing illumination, and camera calibration problems.

Image from camera 1 (some regions are occluded)

Image from camera 2 (not occluded)

Occluded parts of image 1 are now recovered using bootstrapping

Deformable models and topology analysis: In collaboration with Dr. Allili we are trying to develop deformable models mechanisms that take the topology information into account in order to preserve it or to use as a high level knowledge during the tracking.

Up

Some publications

Papers in Refereed Journals

•    Road vectors update using SAR imagery: a snake based method, IEEE Transactions in Geoscience and Remote Sensing, vol. 41, no. 8, August 2003. (PDF )
•    Spatial database updating using active contours for multispectral images: application with Landsat 7, ISPRS Journal of Photogrammetry and Remote Sensing, vol. 57, pp. 346-355, April 2003. ( PDF )
•    Automatic determination of mass functions in Dempster-Shafer theory using fuzzy c-means and spatial neighborhood information for Image Segmentation, Optical Engineering, vol. 41, no. 4, pp. 760-770, April 2002.
•    Dempster-Shafer basic probability assignment based on fuzzy membership functions, Electronic Letters on Computer Vision and Image Analysis, vol 4, no. 1, pp. 1-9, 2004.
•    Algebraic Topology-Based Image Deformation: A Unified Model, To appear in Electronic Letters on Computer Vision and Image Analysis, 2005. (PDF )

Refereed Conference Publications

•    Iterative Markovian estimation of mass functions in evidence theory, 2005. (PDF )
•    A robust Level Set approach for image segmentation and statistical modeling, Proceedings of Advanced Concepts for Intelligent Vision Systems, ACIVS 2004, Brussels, Belgium, September 2004. ( PDF )
•   Neuroanatomy registration using viscous fluid motion: An algebraic-topology based approach, Proceedings of the Third Workshop on Articulated Motion and Deformable Objects, AMDO 2004, pp 76 - 85, Palma de Mallorca, Spain, September 2004. ( PDF )
•   A combined estimation-deformation model for area detection: application to topographic area feature update, Proceedings of Photogrammetric Computer Vision, PCV’02, part A, pp. 181-187, Graz, Austria, September 2002.
•   Automatic Determination of Mass Functions in Dempster-Shafer Theory Using Fuzzy Logic and spatial neighborhood information for Image Segmentation, Proceedings of Vision Interface, VI01, pp. 149-156, Ottawa, Canada, Mai 2001. (PDF )

Up

Teaching

The course that I am teaching in the department of computer science at Bishop's University are

•    CSC 208, Scientific Computing, (Floating-point computation, Nonlinear equations, Numerical integration and differentiation, Curve fitting, FFT and applications)
•    CSC 211, Computer Organization, (Boolean Algebra and Gates, Combinational Circuits, Logic Design, the ALU, Memory, I/O Devices and the Control Unit)
•    CSC 309, Operating Systems, (Device management, Processes and threads, Scheduling strategies, Synchronization, Virtual memory, File Management)
•    CSC 313, Computer Vision, (Binary image analysis, Pattern recognition, Color, Texture, Content-based image retrieval, Motion from 2D image sequences)
•    CSC 312, Image Processing, (Image formation, Image enhancement, Restoration, Feature extraction, Segmentation)
•    CSC 301, Techniques of Simulation, (Declarative Modeling, Functional Modeling, Constraint Modeling)

Up