FBI seeks AI solutions for altered fingerprint detection

On August 24, the FBI CJIS Division posted a request for information regarding current AI capabilities applicable to altered fingerprint detection and matching. Responses are due by 5:00 p.m. Eastern on October 12.

The U.S. Department of Justice (DOJ), Federal Bureau of Investigation (FBI), Criminal Justice Information Services (CJIS) Division, Biometric Services Section, Biometric Identification and Analysis Unit, located in Clarksburg, West Virginia, is requesting information regarding the use of Artificial Intelligence (AI) in the context of altered fingerprint detection and matching.

The FBI’s Next Generation Identification (NGI) System provides the criminal justice and intelligence communities with the world’s largest and most efficient electronic repository of biometric and criminal history information. To access a criminal history record, fingerprints are collected from an individual and submitted to the NGI System for search against ___ fingerprint images from ___ identities. Responses return to contributors within seconds or minutes, depending on the type of transaction. Speed and accuracy are critical components of the NGI System.

The CJIS Division has identified a growing trend in which criminals intentionally alter their fingerprints to defeat identification within the NGI System. Alteration may occur in various ways, including abrading, cutting, applying acid, or performing surgery on the fingertips. These techniques obliterate, distort, or imitate fingerprints of other individuals for the purpose of masking that person’s identity, thus avoiding identification to a criminal history record.

Additionally, unintentional alterations of the friction ridge pattern occur. Individuals whose occupations require frequent contact with chemicals or rough surfaces, such as bricklayers may experience reduced friction ridge detail over time. Certain diseases and medical treatments may also cause fingerprint alterations. Regardless of whether the alterations are intentional or unintentional, accurate identification through the NGI System may be similarly compromised.

Altered fingerprint detection and matching technology have been the topic of numerous scientific papers over the last half decade. Algorithm development efforts have included the capability to detect unaltered and altered fingerprints, and to match altered fingerprints with their unaltered mates.

The next step in the evolution of fingerprint alteration defensive technology is to leverage AI to enable the NGI System to detect and match not only the alteration types the algorithm has been coded to detect and match, but also new alteration types the algorithm has not previously received. As those who seek to avoid identification continue to evolve their alteration techniques, it is critical that the NGI System maintain pace through the ability to learn in real time.

Full information is available here.

Source: FedBizOpps