Engineering Calibrated Biometrics Systems

Abstract

Problem: Performance gap between technology tests and fielded systems: Integration into exiting processes and systems (e.g., isolated watch list systems), Environmental factors, quality, data interchange; Accounting for human behavior and operational controls (intent and tactical variances). Systems level performance predictions are desired: Bayesian models; Human intent and sensor performance; Alternate explanations; Prior probabilities. Systems level engineering method is desired (assume security policy in effect): Determine system boundaries; Define physical environment (e.g., room, facility, vicinity); Sensor model development; Define assessment software, all {Sensor-Software-Application}-tuples; List infrastructure needs; Define collection, alerts, and reporting plans.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Apr 01, 2003
Accession Number
ADA588660

Entities

People

  • Joe Marques
  • Margaret Lyell
  • Nicholas Orlans
  • Paul Lehner

Organizations

  • MITRE Corporation

Tags

Communities of Interest

  • Materials and Manufacturing Processes
  • Sensors

DTIC Thesaurus Topics

  • Application Software
  • Artificial Intelligence Software
  • Bayesian Networks
  • Biometrics
  • Computer Vision
  • Corporations
  • Data Science
  • Engineering
  • Human Behavior
  • Information Science
  • Knowledge Management
  • Models
  • Predictive Modeling
  • Probability
  • Recognition
  • Standards
  • Statistics

Fields of Study

  • Computer science

Readers

  • Enterprise Information Systems Architecture and Joint Command Capability Interoperability Support.
  • Regression Analysis.
  • Theoretical Analysis.

Technology Areas

  • AI & ML
  • AI & ML - Autonomous Systems
  • AI & ML - Bayesian Inference