A synthetic dataset for evaluating soft and hard fusion algorithms

Jacob L. Graham; David L. Hall; Jeffrey Rimland

doi:10.1117/12.884042

23 May 2011 A synthetic dataset for evaluating soft and hard fusion algorithms

Jacob L. Graham, David L. Hall, Jeffrey Rimland

Proceedings Volume 8062, Defense Transformation and Net-Centric Systems 2011; 80620F (2011) https://doi.org/10.1117/12.884042
Event: SPIE Defense, Security, and Sensing, 2011, Orlando, Florida, United States

Abstract

There is an emerging demand for the development of data fusion techniques and algorithms that are capable of combining conventional "hard" sensor inputs such as video, radar, and multispectral sensor data with "soft" data including textual situation reports, open-source web information, and "hard/soft" data such as image or video data that includes human-generated annotations. New techniques that assist in sense-making over a wide range of vastly heterogeneous sources are critical to improving tactical situational awareness in counterinsurgency (COIN) and other asymmetric warfare situations. A major challenge in this area is the lack of realistic datasets available for test and evaluation of such algorithms. While "soft" message sets exist, they tend to be of limited use for data fusion applications due to the lack of critical message pedigree and other metadata. They also lack corresponding hard sensor data that presents reasonable "fusion opportunities" to evaluate the ability to make connections and inferences that span the soft and hard data sets. This paper outlines the design methodologies, content, and some potential use cases of a COIN-based synthetic soft and hard dataset created under a United States Multi-disciplinary University Research Initiative (MURI) program funded by the U.S. Army Research Office (ARO). The dataset includes realistic synthetic reports from a variety of sources, corresponding synthetic hard data, and an extensive supporting database that maintains "ground truth" through logical grouping of related data into "vignettes." The supporting database also maintains the pedigree of messages and other critical metadata.

Citation Download Citation

Jacob L. Graham, David L. Hall, and Jeffrey Rimland "A synthetic dataset for evaluating soft and hard fusion algorithms", Proc. SPIE 8062, Defense Transformation and Net-Centric Systems 2011, 80620F (23 May 2011); https://doi.org/10.1117/12.884042

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available