You are here

Topological Signal Analysis for Multi-Modal Data Analysis

Award Information
Agency: Department of Defense
Branch: Air Force
Contract: FA8750-16-C-0220
Agency Tracking Number: F16A-T12-0028
Amount: $149,822.00
Phase: Phase I
Program: STTR
Solicitation Topic Code: AF16-AT12
Solicitation Number: 2016.0
Timeline
Solicitation Year: 2016
Award Year: 2016
Award Start Date (Proposal Award Date): 2016-08-02
Award End Date (Contract End Date): 2017-04-20
Small Business Information
636 Rock Creek Road
Chapel Hill, NC 27514
United States
DUNS: 078652742
HUBZone Owned: No
Woman Owned: Yes
Socially and Economically Disadvantaged: No
Principal Investigator
 Paul Bendich
 (919) 660-2811
 bendich@math.duke.edu
Business Contact
 John Harer
Phone: (919) 448-7871
Email: geomdata@gmail.com
Research Institution
 Duke University
 Keith Hurka-Owen
 
Office of Research Administration 2200 W. Main Street, Suite 820
Durham, NC 27705
United States

 (919) 681-8687
 Nonprofit College or University
Abstract

Extracting information from different modalities of data is one of the greatest challenges in this era of exponential data growth, as well as one of the most promising horizons for the development of new capabilities for military, commercial, and scientific applications. Virtually all areas of defense, industry, and science are becoming increasingly dependent on the ability to analyze multimodal data, and consider the difficulties of this process as one of the most important bot- tlenecks preventing progress. This proposed project addresses one aspect of this challenge by developing methods to detect targets of interest in a variety of situations, using multi-modal sensor data from heterogeneous ISR data sources. We look to not only associate that data, but to do so with appropriate measurements of accuracy in order to use these joint data correctly. Joint distributions on multi-modal data can then provide meaningful probabilities that allow us to deal with conflicts or ambiguities from different sources, and deal more accurately with missing data in one or more sources due to noise, sensor failure or occlusion.

* Information listed above is at the time of submission. *

US Flag An Official Website of the United States Government