Rapid Anomaly Detection and Tracking via Compressive Time-Spectra Measurement

Abstract

This report describes the results of a two-year Phase II STTR program undertaken by InView in partnership with Rice University to develop a compressive sensing based imager with high speed anomaly detection capability. During this period, novel multi-channel compressive sensing measurement patterns and detector geometries were developed along with statistical methods for performing anomaly detection on data in the compressed domain. Detection performance was analyzed using receiver operating characteristic and precision recall curves. New concepts, algorithms and techniques were simulated and tested experimentally using specially modified compressive sensing cameras under laboratory conditions. It was found that the number of measurements needed for anomaly detection was far less than the number of measurements needed for compressive imaging, validating high speed system operation.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Feb 12, 2016
Accession Number
AD1008565

Entities

People

  • Kevin F. Kelly
  • Lenore Mcmackin
  • Matthew A. Herman
  • Tyler Weston

Tags

Communities of Interest

  • Advanced Electronics
  • Energy and Power Technologies
  • Human Systems
  • Materials and Manufacturing Processes
  • Sensors

DTIC Thesaurus Topics

  • Algorithms
  • Anomaly Detection
  • Change Detection
  • Compressed Sensing
  • Computational Science
  • Computer Vision
  • Data Acquisition
  • Detection
  • Detectors
  • Focal Plane Arrays
  • Machine Learning
  • Mathematical Analysis
  • Mathematical Models
  • Modulation
  • Optical Modulators
  • Pattern Recognition
  • Two Dimensional

Readers

  • Computer Vision.
  • Research Science/Academic Research
  • Structural Health Monitoring of Composite Structures.