Computational Science and Intelligence Lab: Projects
Current Projects | Previous Projects
Listing of some of our projects - Not all projects are listed here
Current Projects
Optimized Multi-Algorithm Systems for Detecting Explosive Objects Using Robust Clustering and Choquet Integration, National Science Foundation Program
Dates: September 2007 - August 2010
Collaborators:University of Louisville
Description: This project will develop multi-algorithm systems that detect explosive objects from sensor data obtained through detecting tools such as ground penetrating radars and hyperspectral imagers. Analysis of context will be used as a precursor to detection of landmines.
Students on Project: Jeremy Bolton
GSTAMIDS: Ground Standoff Minefield Detection System
Collaborators: University of Louisville, University of Missouri - Columbia, Duke University, Niitek, Institute for Defense Analyses
Description:Development of algorithms for classification and fusion problems using vehicle-mounted ground penetrating radar data.
Students on Project: Jeremy Bolton, Xuping Zhang, Ryan Busser, Ken Watford
AMDS: Autonomous Mine Detection Sensors Program
Collaborators:University of Missori - Columbia
Description:Develop algorithms for multi-sensor robotic mine detection system.
Students on Project: Ganesan R, Seniha Esen Yuksel, Xuping Zhang
HSTAMIDS: Handheld Standoff Mine Detection System
Description: Algorithm development for ground-penetrating radar and metal detector data collected using a hand-held sensor.
Students on Project: Raazia Mazhar
Previous Projects
Listing of some of our projects - Not all projects are listed here
Multi-University Research Initative: Science of Land Target Spectral Signatures
Dates: August 2003 - May 2008
Collaborators: Georgia Institute of Technology, University of Maryland, University of Hawaii, Rochester Institute of Technology, Clark Atlanta University
Description: Performing hyperspectral image analysis and investigating methods for algorithm fusion with applications to mine field detection. Investigating endmember detection algorithms and context based classification. Implementing methods for LWIR vegetation detection for false alarm reduction. Applying non-linear aggregate techniques using the Choquet Integral and random set theory for improved fusion and classification results. MURI led by Georgia Institute of Technology
Students on Project: Jeremy Bolton, Alina Zare, Andres Mendez-Vazquez
