Next in this series on physical and behavioral biometric technologies, Ravi Das continues on the topic of facial recognition technology. In the previous article the author described the various commercial and governmental applications of this technology. Here he does a deep dive into exactly how facial recognition systems work.
Facial recognition is a biometric technology that most people know exists. For example, every individual has a face, and just like the fingerprint, the face has long been used to verify and identify criminals, wanted suspects, and terrorists. Probably one of the best examples of facial recognition is the “WANTED” photos at the post office, as well as the facial images on the “Wanted” section of major law enforcement websites.
But, unlike the other biometric technologies being used today, facial recognition is especially subject to privacy rights issues and claims of civil liberties violations. The main reason for such objections is that facial recognition technology can be used very covertly, without the knowledge or consent of the individuals the system is trying to track down.
Two key advantages of facial recognition technology are that it can be used for both verification and identification scenarios, and it can be deployed for heavy duty usage. For instance, facial recognition technology is used frequently in the e-Passport infrastructures of many nations around the world and in large scale identification applications at the major international airports, especially to hunt down suspects on the terrorist watch lists.
Factors that can “trick” facial recognition technology
Facial recognition has its fair share of technological flaws. For example, if a facial recognition system were to capture the image of an individual who is grossly overweight, and then capture another image of the same person after he experienced massive weight loss, the facial recognition system would be unable to make a positive match. In other words, the system can be very easily spoofed in this way.
Other characteristics that can trick facial recognition technology are the presence and subsequent removal of facial hair, aging, as well as the presence or absence of other facial accessories, such as hats, sunglasses, switching from contact lenses to eyeglasses, and so on.
Facial Recognition Technology: How It Works
Facial recognition technology relies on the physical features of the face (see Figure #1), which are determined by genetics.
Figure 1
Facial recognition systems of today focus on those parts of the face which are not as easily prone to the hurdles described above. These regions of the face include:
- The ridges between the eyebrows
- The cheekbones
- The edges of the mouth
- The distances between the eyes
- The width of the nose
- The contour and the profile of the jawline
- The chin.
This technology can be deployed either as a fully automated system, or as a semi-automated system. With the former, no human interaction is required; all the verification and identification decisions are made by the technology itself. With the latter, a degree of human intervention is required, and this is the preferred method for deploying a facial recognition system. Given some of the serious obstacles this technology still faces, it is always better to err on the side of caution by having an actual human being involved in rendering a verification or identification decision.
The methodology used to capture the raw images of the face is very different than other biometric technologies. Although facial recognition is a contactless technology, the image capture processes are far more complex.
- Collecting raw images of an individual who is knowingly having his face “scanned” requires cooperation from that person.
- In the case of covert facial recognition surveillance methods, the individual unknowingly has their face captured using facial recognition technology implanted in a Closed Circuit Television (CCTV) camera system.
Once the raw images are collected by the camera, the data is then either aligned or normalized to help refine the raw images at a more granular level. The refinement techniques involved include adjusting the face to be in centered in the pictures which have been taken, and adjusting the size and the angle of the face so that the best unique features can be extracted and later converted to the appropriate verification and enrollment templates.
All of this is done via mathematical algorithms. As mentioned previously, facial recognition technology is countered by several major obstacles, especially during the raw image acquisition phase. These obstacles include:
- A lack of subtle differentiation between the faces and other obstructive variables in the external environment.
- Various facial expressions and poses in subsequent raw image captures.
- Capturing a landmark-orienting feature such as the eyes.
Read on to learn what is being developed to resolve challenges such as those described above.
Advances in 3-Dimensional imaging
To help compensate for the current obstacles, much research and development has been done in what is known as 3-Dimensional imaging. In this technique, a shape is formed and created, and using an existing 2-Dimensional image, various features are created. This results in a model which can be applied to any 3-Dimensional surface and can also be used to compensate for the above-mentioned challenges.
However, it should be noted that 3-Dimensional facial recognition systems are not yet widely deployed in commercial applications, because the technique is still considered to be in the research and development phases. Right now, it is primarily 2-Dimensional facial recognition systems which are used in the commercial market. 3-Dimensional facial recognition systems are only used as a complement to the 2-Dimensional systems, which dictate higher imaging requirements, and the capture environment is much more challenging.
Up Next: The next article in this series will continue to explore facial recognition technology, by describing the various facial recognition models and techniques—and the scientific and mathematical factors that support them
Ravi Das is a Cybersecurity Consultant and Business Development Specialist. He also does Cybersecurity Consulting through his private practice, RaviDas Tech, Inc. He also possesses the Certified in Cybersecurity (CC) cert from the ISC2.
Visit his website at mltechnologies.io
