US Army calls for proposals on face biometric base access and ABIS data cleansing

Two new calls for proposals are aimed at implementing face biometrics at U.S. Army bases and installations, and a machine learning solution to overhaul the DoD’s widely-used ABIS database.

The U.S. Army is accepting proposals for an enhanced face biometrics solution that would allow base access using drive-thru gates. A similar program is currently being piloted at the Redstone Arsenal in Alabama. Program implementation will be carried out in three phases and focuses on an enhanced camera system that allows security to capture face biometrics of multiple passengers regardless of light and weather-related visibility conditions.

Phase one includes system design and development of an agnostic and scalable platform that can support third-party systems and access control solutions, data leverage capability for existing biometric data, mobility, a programmable field of view, resolution requirements, physical size requirements, and ability to authenticate multiple users.

The second phase consists of a demonstration of the system in question in a real-world scenario. Specific requirements are face biometrics capture at speeds of 5 to 10 mph, 100 percent accuracy, processing speed, and authentication capabilities. Phase three will test whether the system can be deployed to other military access control points (ACPs).

Another call for proposals was released for the Department of Defense’s Automated Biometric Identification System (ABIS). Specifically, the announcement seeks submissions for machine learning software to conduct data cleansing of the widely used biometric database. This solution would allow biometric face, iris, palm, and latent image data to be cleansed to improve the system’s accuracy and performance. The program will be implemented in three phases over a 3 to 5-year window.

The announcement further explains, that the DoD ABIS, as the authoritative data base, is required to accept submissions from across the department, but much of the data comes from legacy systems that are nearing obsolescence and do not have adequate data quality safeguards. The database has a large number of biometric records with errors or missing data, therefore.

“This poor quality data uses valuable computing resources as well as results in a higher number of ‘yellow’ matches which require a human examiner to review the request as well as to manually determine if there is or is not a match in the biometric data within the data base and the request,” DoD explains in the announcement.

Article Topics

ABIS  |  access control  |  biometric database  |  biometric identification  |  biometrics  |  facial recognition  |  machine learning  |  tender  |  U.S. Army