Biometric Identity Verification: The Difference Between Facial Recognition and Facial Comparison Technology

Conor Hickey, June 6, 2021

Biometric identity verification technologies, such as facial recognition and facial comparison, are now commonplace in our daily lives. Every day we use fingerprint technology or face recognition to unlock our smartphones using FaceID, and at airports across the world, many of us now move seamlessly through passport control using ID document verification, artificial intelligence, and facial comparison technologies to validate our identity.

While some groups such as the ACLU have raised concerns around privacy, civil liberties, and human rights if this technology were to be abused, it also presents commercial and cybersecurity benefits alongside other biometric technologies. Other cities including Boston and San Francisco have gone so far as to ban the use of the technology while New York, the FBI, and the Department of Homeland Security have strict rules around its use. A face detection and identification system is just a tool. Here we'll parse the definitions distinguishing the use of facial recognition technology and facial comparison.

The Difference Between Facial Recognition and Facial Comparison

The terms facial recognition and facial comparison refer to two different ways of using the same technology. Both of these new technologies are used to verify that a source image of a human face matches another image, either stored or supplied.

Facial recognition assumes that there are recognition databases of multiple images that the source image recognition data can be compared against – the source image is then recognized in the store of images. Facial comparison assumes that two images of the same person are being compared.

Digital Account Opening: How Banks Can Transform & Protect The Customer Journey

Digital Account Opening: How Banks Can Transform & Protect The Customer Journey

Learn key trends, best practices, and technologies to overcome the challenge of creating a fully digital account opening process, including facial biometrics, e-signature, and machine learning-based fraud analytics in this white paper.

Download Now

Use of Facial Recognition Software

Facial recognition is commonly used in law enforcement where a database of images of known criminals can be searched, and the relevant authorities notified on a match.

Face recognition technology is also commonly used for enterprise-wide security. In facial recognition security systems, a database of employee facial images can be searched, and the employee is only allowed access to a restricted area (either physical or virtual) on a successful match.

We can look to the proliferation of facial recognition solutions in China to understand how the technology might be used in the future.

In China, personal information is available through government sources and facial recognition is already being used for a variety of applications, from law enforcement agencies to food payments. The police departments in China have tested smart glasses that implement a face recognition system to scan peoples’ faces. These images are then compared against a database with other known criminal mugshots and outstanding warrants. In the consumer sector, 300 branches of KFC now allow customers in China to ‘Pay by smile’, and to deter toilet paper thieves, toilet paper is dispensed in a public toilet in Shanghai by a system based on facial recognition technology.

Use of Facial Comparison Software

Facial comparison is typically used where a trusted source image of the user is available to compare a real-time image against. The trusted source picture is usually extracted from an authenticated identity document such as a passport. For this reason, document verification and facial comparison are commonly used together.

One common use case for facial comparison is automated passport control (‘e-gates’). The process uses ID document verification and facial comparison together to verify the authenticity of the ID document and the presence of the genuine owner of the passport in real time. During this process a trusted source image of the passport holder is compared with the real-time photo of the person trying to get through the gate.

Here’s how facial comparison is used to prove that a person is present at passport control:

  1. The user presents their passport for authentication
  2. An image of the user’s photograph is extracted from their passport
  3. The image is used as a trusted source image
  4. The automated gate takes a photo of the user
  5. Facial comparison compares the trusted source image (extracted from the passport) with the photograph
  6. If the trusted source image and the photograph match, the user is allowed through

Facial comparison is also used by financial institutions to verify the identity of an applicant or unknown user when that user is not physically present, for example during digital account opening.

Here’s how facial comparison is used to verify the identity of an applicant and prove that an applicant is present during digital account opening:

  1. Document verification is used to verify the authenticity of an applicant’s passport, ID card, or driver’s license.
  2. Once the authenticity of that ID document is confirmed, the applicant is asked to take a selfie using their handheld device.
  3. Facial comparison technology compares the selfie image with the image from the verified ID document to prove that the verified person is genuinely present during the account opening process.

When facial comparison is used for digital account opening, it uses advanced facial recognition algorithms to extract biometric data from a facial image — distilling facial features (such as the position and size of a person’s eyes relative to each other) into a standardized dataset. Comparing two datasets can determine whether two images are from the same individual. If one image is from a pre-verified source (e.g. a passport or ID card verified using document verification) and the second image is a real-time image taken from the applicant at the time of their application, facial comparison can be used to prove their presence and recognition as the applicant. In a post-pandemic world where remote processes are vital, images of individuals are readily available on social media, large-scale fraud attacks steal personal data right from the user's Apple iphone or Microsoft Android device, facial recognition systems are going to become all the more important for ID verification and cyber security.

Conor Hickey is the Head of Solution Architecture for OneSpan’s Secure Agreement Automation product in EMEA, specializing in the digitization of client on-boarding processes in the financial services sector.