How does skin-detection identification software work?

By now, you may be aware that biometrics is one of the fastest growing segments of the security industry and is primarily used to grant access to services. Some of the familiar techniques for identification are facial recognition, fingerprint detection, handwriting verification, hand geometry, retinal and iris scanner. Of these, face and fingerprint recognition have been the ones that have rapidly developed.  This is due to their widespread use to unlock our smartphones, as these methods are user friendly and much more secure than a password alone.

Yet, despite the progress made with this technology, there have been cases where it has been fooled by 3D printed masks. This was particularly alarming as it proved that some facial recognition systems can authenticate users even when they are not alive.

As such, facial recognition cameras have come to rely on ‘liveness detection’ in order to grant access to a service. Here, algorithms analyse the images or videos and decide whether they come from a live person or a fake. Methods used are motion and/or texture analysis of the face, as well as artificial intelligence so the system can continue to adapt to changes in your expression, weight, hairstyle, and accessories, and recognise your face more quickly. For example, even if you wear a scarf or grow a beard, with AI, the system will learn to recognise you.

How does skin-detection work?

Using “Beam Profile Analysis” technology, three data streams are taken from a single camera system, a 2D image, a 3D depth map and most uniquely – material classification.

Through this fundamentally new approach to security, it is now possible to combine standard facial recognition algorithms from any third party with the unique ability to sense “live skin.” The technology works by using infrared beams to map out the fine structures of the face, creating a 3D representation to verify the user. It then analyses how this infrared light is backscattering light on different surfaces.

Put simply, the way skin reflects light is different from the way a silicon mask or a photograph would for example. Importantly, the software works on all skin colours and genders as the light scatters the same way despite physical differences between people. This is particularly relevant to companies developing facial recognition software, as in 2019 a study by the US government discovered that facial recognition systems misidentified people of colour more often than white people.

Early camera prototypes use a small Raspberry Pi computer that works with the LG phone’s USB-C port, but the actual camera array should use only the phone’s internal processor, and rely on similar infrared dot projectors and camera sensors like existing face-scanning phones, though aligned in a way that will work with different algorithms. For example, Apple’s Face ID already projects 30,000 infrared dots onto your face to help create a 3D map of your face when authenticating you. This means the technology will not be very expensive to implement.

By 2021, the company developing the skin-detection hopes to have its facial recognition system in Android and Windows devices that run on Qualcomm’s Hexagon processor. Some devices that run on this processor include:

  • Samsung Galaxy S11, Galaxy S11 Plus, and Galaxy S11e
  • Google Pixel 5 and Pixel 5 XL
  • Sony Xperia 2
  • LG G9 and LG V60

The technology also has the potential to be used in many more cases than smartphone authentication. For instance, it could also help computer vision systems in autonomous vehicles or be used in factories or warehouses where robots need to search and collect particular items by material.


The reason why facial recognition has become so popular amongst smartphone makers is the level of security and convenience it provides to their customers. Apple have said for example that the chance of a random person being able to unlock your phone with Face ID is 1 in 1,000,000. Touch ID by comparison was 1 in 50,000, so it’s a significant improvement. With skin-detection software included, this number will get even lower, meaning that hackers will find it much harder to gain access to your devices.

As most highly publicised breaches are attributed to weak or absent authentication (vulnerable passwords, unlocked laptops or wireless networks), these new verification methods will help protect against unauthorised access.

Let us know if you have any questions about skin-detection or biometrics in the comments below.