Trends in Multimodal Biometric Identification Systems
Posted on May 21, 2021 by Chuck Collins
Numerous physical characteristics exist in each of us that enable law enforcement personnel to accurately identify criminal suspects. The most important are a person’s fingerprints, palm prints, facial features, irises and DNA. Individually these characteristics—which are unique to everybody—are called modalities. Three other significant modalities are a person’s gait, the timber of their voice and their vascular system—the manner in which veins and arteries grow also is unique to each individual—though these are emerging modalities.
Another important use for the data associated with these modalities concerns access control. This can pertain to physical locations as well as to computer and communications networks and systems. In a prison, for example, modalities can be used to restrict or allow prisoner movement and access to specific areas, such as the commissary, library, pharmacy or exercise yard.
Multimodal Biometric Identification Systems (MBIS)
Capturing, matching, archiving and sharing data associated with these modalities is accomplished via multimodal biometric identification systems (MBIS). Over time, background databases can be created that are extremely useful. For instance, let’s say that a person is arrested, resulting in their fingerprints being captured. Years later, the person is arrested again, and the fingerprints captured during that incident can be compared with the background database, which reveal that the person is a repeat offender. This is very useful information for law enforcement personnel and prosecutors to have at their fingertips—for example, if the person tried to use an alias for either arrest, they immediately would be exposed during the second arrest.
MBIS-generated data also comes in handy for civilian purposes. Someone seeking access to a vulnerable population—e.g., a baseball coach, scout leader, schoolteacher, school bus driver or assisted-living facility worker—might need to undergo a background check that would involve one or more of the aforementioned modalities. The last thing that any organization dealing with a vulnerable population would want to do is insert a felon into that environment, for numerous reasons, and the data generated and stored by an MBIS helps to prevent that from occurring.
The History Behind the First MBIS
The first MBIS emerged in the 1980s. Prior to that, gathering and comparing fingerprints—the only modality captured for decades—was a time-consuming manual process. Fingerprint identification was pioneered by Sir Francis Galton, an anthropologist who was a cousin to Charles Darwin. Galton determined that the fingerprint was a highly reliable identifier for two reasons: no two fingerprints are exactly alike, even in identical twins; fingerprints remain constant through a person’s lifetime. He also identified the three most common fingerprint types—loop, whorl, arch—and these classifications still are used today.
Galton published his findings in 1892; shortly thereafter, Sir Edward Richard Henry used those findings to develop a fingerprint-classification system that bears his name, and which became the dominant system used in law enforcement. In 1901, Scotland Yard leveraged the Henry Classification System to begin compiling fingerprint data, and two years later the New York City Police Department, New York State Prison System and Federal Bureau of Prisons followed suit.
In 1910, fingerprints were used for the first time to convict a murderer, which occurred in Illinois. The verdict was appealed but eventually upheld by the state’s Supreme Court. These court cases established the reliability of fingerprint evidence.
While the Henry Classification System was surprisingly efficient, it still was laborious—plus it only involved one modality. Leveraging the theory that more is better, an MBIS is a vast improvement upon reliance on fingerprints as the sole identification evidence in a criminal case.
The First Multistate, Shared Automated Fingerprint Identification System (AFIS)
However, such systems, when they first emerged roughly three decades ago, were prohibitively expensive. For that reason, six original (now eight) western states formed the Western Identification Network (WIN) as a means of pooling their resources to implement the first multistate, shared automated fingerprint identification system (AFIS). WIN, which is an MCP client, still exists today as a non-profit consortium of state and local law enforcement agencies that collectively maintains and shares a multistate MBIS. This was a remarkable achievement on several levels, but especially when one considers that it took the FBI many years to implement its first multistate AFIS.
Today's MBIS Trends
Eventually, the AFIS capability expanded into today’s MBIS. And, like every technology solution, MBIS have evolved over the years. The following are a few of the most important current trends:
- Arguably the most important is the movement away from a hardware-centric environment. Prior to this, comparing fingerprints to the background database involved a lot of computer hardware, i.e., microchips and processors—there would be many racks of matchers. Today MBIS largely are software-centric, which is good news for law enforcement and justice agencies because software is now faster, less expensive, and more manageable compared with hardware.
- A shift to hosted solutions increasingly is gaining traction. There are several advantages to hosted solutions. There is no hardware or software for the agency to procure, install, operate or maintain. While a hosted solution creates a monthly recurring expense, the long-term cost usually is far less than what would be incurred with an agency-hosted solution.
Also, such expenses fall into the category of operational expenses, which typically are easier to justify compared with capital expenses, i.e., system procurement and installation, from a budgeting perspective. However, the most important advantage of hosted solutions is that they provide the agency with higher levels of availability and failover compared with the agency-hosted solution and allow the agency to focus on their law-enforcement missions rather than expending resources to maintain technology.
- Another trend associated with the hosted approach involves virtual desktop infrastructure (VDI), which removes the biometric workstation from the equation. Users simply log into the system—which is hosted at another facility that often is hundreds, even thousands of miles away—and all of the files to which they need access are available to them via a secure, virtual private network (VPN) connection.
The big advantage to VDI—besides the cost savings that are similar to those offered by hosted solutions—is that whenever a vendor needs to execute an update to the workstation, it is done at the hosted facility and is immediately available the next time each user logs on. In contrast, past methods included remote downloads and configurations, or even having the vendor physically visit the agency’s site and touch each workstation individually, a time-consuming and operationally challenging approach. This is especially true of agencies that are located in remote locations. This trend today is at a nascent stage, but is expected to be the norm going forward.
MCP has numerous subject-matter experts who are eager to chat with you about MBIS and the best approaches to implementing them—please reach out.
Chuck Collins is MCP’s vice president of public safety. He can be emailed at ChuckCollins@MissionCriticalPartners.com.