Facial recognition is becoming an increasingly essential layer within federal security. Over the past decade face recognition has matured to the point that it should now be considered an essential part of any government threat prevention technology stack. By using face recognition, government organizations identify persons of interest in real time, mitigate risks and increase operational effectiveness. By fully automating the manual identification process, face recognition saves agencies valuable time, while empowering them to leverage critical, life-saving data in the context of missions.
When evaluating a face recognition system, government agencies should be cognizant of the following five key success factors:
We’ve found that speed is an essential metric for enterprise-level deployments and tends to be a critical factor when agencies are evaluating technologies. It’s vital that a face recognition can quickly match images against massive databases in order to establish suspects’ identities in real time. FaceFirst just improved our system to now search and match against 25 million faces per second. This enables our system to deliver matches in real time even with massive databases.
In addition to match speed, accuracy is the most important success factor for face recognition. Systems that produce a high degree of false positives or false negatives are just going to waste agents’ valuable time. Fortunately, face recognition has grown significantly more accurate over the past several years. As an example, FaceFirst now uses approximately 150,000 points of reference on a face when establishing identity; this is approximately 5 times more than the Apple iPhone X uses to authenticate users. By using more points of identification on a face, it helps ensure match accuracy.
Government agencies accumulate tons of data about potential suspects. But far too often, this data remains siloed and cannot be accessed in real time. During critical missions, seconds can sometimes spell the difference between success and failure (if not life and death.) That’s why it’s so important to invest in a system that can instantly alert the right people at the right time with contextual information. As an example, FaceFirst set up our system to allow our customers to configure exactly who sees alerts. Alerts can be sent to individual agents or agent groups, both in the field or at control centers.
Since agencies use a variety of threat prevention technologies, it’s often imperative to use a face recognition system that is interoperable with existing and third-party technologies. Look for systems that can be deployed as a wholly contained solution, or integrated into existing video management systems (VMS), intelligence platforms, analytics platforms and other solutions.
Data security is another critical factor when evaluating face recognition systems. Government organizations and end users need to be confident that data is secure. At FaceFirst, we’ve baked several data security features into our platform. Here are some examples of some ways that our platform keeps critical data safe:
- Encryption – biometric data is encrypted at rest and during transmission
- Data breach precautions – biometric templates stored within the FaceFirst system cannot be converted back into a face image in the case of a data breach
- Data purging – biometric data can be purged according to strict timetables
- Checks and balances – Role hierarchies ensure only authorized individuals can approve and view enrollment images within the FaceFirst system
For a more in-depth look at how FaceFirst can help government agencies achieve objectives, download our free whitepaper.