Discrimination, Identification, and Tracking of Unresolved Targets Using Tomographic Integration of Multiplex Sensor Data

Abstract

The objective of this research is the development of a low cost sensor system with applications in biometric tracking and authentication. We have demonstrated the ability to classify and track human motion using low cost, off-the-shelf pyroelectric detectors. We have designed fieldable sensor units that are anonymous and can perform multiple tasks such as biometric classification, multiple object tracking, and camera pointing. We have demonstrated the ability to track two individuals and demonstrated the ability to recognize multiple humans with an accuracy of 86%. We have integrated the units through wireless communication to allow the creation of a cluster field of detectors.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Dec 20, 2006
Accession Number
ADA470868

Entities

People

  • David Brady

Organizations

  • Duke University

Tags

Communities of Interest

  • Sensors

DTIC Thesaurus Topics

  • Algorithms
  • Authentication
  • Computational Complexity
  • Detection
  • Detectors
  • False Alarms
  • Hidden Markov Models
  • High Resolution
  • Homeland Defense
  • Homeland Security
  • Identification
  • Identification Systems
  • Infrared Detectors
  • Markov Models
  • Person Tracking
  • Recognition
  • Warning Systems

Readers

  • Distributed Systems and Data Platform Development
  • Image Processing and Computer Vision.
  • Radar Systems Engineering.