SMART

Space-based Machine Automated Recognition Technique
SMART

Intelligence Value

SMART automates broad-area search of multi-source satellite imagery to detect, monitor, and characterize the progression of anthropogenic or natural processes, such as heavy construction or crop growth. By augmenting the manual imagery analysis process with global-scale image processing and machine learning, SMART will provide timely discovery and robust monitoring of man-made and natural change.

Summary

Manual exploitation methods do not scale with the volumes of data available from single individual sensors, and they fail altogether at the problem of simultaneously analyzing data from past, current, and future space-based systems. With the growing quantity and diversity of imagery collected, the Intelligence Community is seeking novel methodologies to improve the analysis process and efficiently distill the data into actionable intelligence.

SMART innovations in data fusion and automated reasoning techniques will enable large-scale monitoring of both man-made and natural change with unprecedented temporal resolution and area coverage, erasing strategic surprise. Harmonization ensures calibration, correction, and georegistration, which allows the creation of a virtual constellation, providing the necessary coverage and temporal resolution for many intelligence problems. Subsequently, machine learning and reasoning activities across spatial, spectral, and temporal features deliver automated sense-making against the harmonized data, enabling global alerting for change of interest.

The SMART program will use detection and monitoring of heavy construction as an initial use case and investigate the transferability of the approach to other forms of natural and anthropogenic change. The ability to accurately characterize the temporal stage of dynamic processes in an automated fashion will validate the mission utility of SMART’s harmonized multi-source imagery and machine learning system.


Contact Information

Program Manager

Dr. Jack Cooper

jack.cooper@iarpa.gov

301-243-2033

Research Area(s)

Activity Recognition, Broad Area Search, Data Harmonization, Machine learning , Time-series analysis

Related Program(s)

Broad Agency Announcement (BAA)

Link(s) to BAA

IARPA-BAA-19-04

Solicitation Status

CLOSED

Proposers' Day Date

May 29, 2019

BAA Release Date

February 7, 2020

BAA Question Period

February 7, 2020 — March 6, 2020

Proposal Due Date

Monday, 18 May 2020

Testing and Evaluation Partners

  • Johns Hopkins University Applied Physics Lab
  • MITRE
  • National Aeronautics and Space Administration Goddard Space Flight Center
  • Savanah River NationalLaboratory
  • United States Geological Survey

Prime Performers

  • Accenture Federal Solutions
  • BlackSky Geospatial Solutions
  • Kitware Inc.
  • Systems & Technology Research
  • Applied Research Associates
  • ASTRA
  • Intelligent Automation, Inc