Fooling Computer Vision Classifiers with Adversarial Examples

Award Information
Agency: Department of Defense
Branch: Navy
Contract: N68335-18-C-0805
Agency Tracking Number: N182-127-0136
Amount: $124,963.00
Phase: Phase I
Program: SBIR
Solicitation Topic Code: N182-127
Solicitation Number: 2018.2
Timeline
Solicitation Year: 2018
Award Year: 2019
Award Start Date (Proposal Award Date): 2018-10-15
Award End Date (Contract End Date): 2019-04-18
Small Business Information
200 TURNPIKE ROAD, CHELMSFORD, MA, 01824
DUNS: 796010411
HUBZone Owned: N
Woman Owned: N
Socially and Economically Disadvantaged: N
Principal Investigator
 Tyson Lawrence
 (978) 856-4158
 tlawrence@tritonsystems.com
Business Contact
 Collette Jolliffe
Phone: (978) 856-4158
Email: cjolliffe@tritonsystems.com
Research Institution
N/A
Abstract
Triton Systems Inc. propose to create adversarial image techniques that will both fool U.S. classifiers and can applied in operationally relevant physical modification kits to military objects. We will develop a rapid process to generate adversarial images, translate them into physical modification kits and demonstrate a reliable system to camouflage the target object independent of classifier location and type. Our developed system will be used to test U.S. Navy deep neural network computer vision classification systems and to develop an effective counter defense against these physical-world adversarial example attacks.

* Information listed above is at the time of submission. *

Agency Micro-sites

SBA logo
Department of Agriculture logo
Department of Commerce logo
Department of Defense logo
Department of Education logo
Department of Energy logo
Department of Health and Human Services logo
Department of Homeland Security logo
Department of Transportation logo
Environmental Protection Agency logo
National Aeronautics and Space Administration logo
National Science Foundation logo
US Flag An Official Website of the United States Government