Facial recognition has become common on smartphones and laptops, but they’re not all equally reliable or even secure.
Face Recognition On Smartphones
Face recognition on smartphones: is it safe and should you use it?
Face recognition on personal use devices such as smartphones and laptops is becoming increasingly common. The concept looks downright futuristic on the surface: Just a few years ago, most of us used a PIN, password or pattern to unlock our devices.
However, despite the rise in technology, you should know that not all facial recognition implementations are created equal. Some techniques are inherently more secure than others, while others offer optional settings to reduce the chance of being tricked by an intruder.
With that in mind, let’s take a look at the different types of facial recognition techniques in use today. We’ll also discuss the security of the feature later and whether or not you should enable it on your devices.
Face recognition with camera
As the name suggests, this technology relies on the front cameras of your device to recognize your face. Virtually every Android smartphone has had this feature since the release of Android 4.0 Ice Cream Sandwich in 2011. This was before fingerprint sensors were as common as they are today, making them the first biometric unlock option.
How it works is pretty simple: when you first turn the feature on, your device will ask you to take pictures of your face, sometimes from different angles, and then use a software algorithm to extract and save your facial features. Every time you try to unlock your device, a live image from the front camera is compared with the reference data.
The accuracy depends mainly on the software algorithms used, so the system is far from perfect. Things get even more complicated when devices have to take into account variables such as different lighting conditions, changes in appearance, and the use of facial accessories such as glasses. , Jewelry and even masks.
While Android offers facial recognition APIs, smartphone manufacturers have also developed custom solutions over the years. Overall, the goal was to improve the device’s detection speed without sacrificing accuracy too much. However, some implementations went too far and could be tricked into accepting a photo.
Infrared Face Recognition
Although most devices already have a front camera, infrared face recognition requires additional hardware. However, not all infrared-based facial recognition solutions are created equal.
The first type of infrared-based face detection takes a two-dimensional photo of your face, similar to the previous method, but in the infrared spectrum. The main benefit is that infrared cameras don’t have to illuminate your face well and can even work in low light environments. They are also much more resistant to attempted infiltration because infrared cameras use thermal energy or heat to create an image.
Take a look at the following image from Microsoft showing how infrared cameras view photos and displays.
Two-dimensional infrared face recognition can be found nowadays mostly on high-end laptops under the umbrella of Windows Hello. This includes Microsoft’s Surface devices and other business-oriented laptops. Aside from that, you can also get external webcams like the Logitech Brio 4K that include IR cameras that are Windows Hello certified.
While 2D infrared face recognition is way ahead of traditional camera-based methods, there is an even better way to go. For example, Apple’s Face ID uses a number of sensors to capture a three-dimensional representation of your face. a floodlight and a point projector to project thousands of tiny invisible points onto your face. Then an infrared sensor measures the arrangement of the points and creates a depth map of your face.
3D systems have two advantages: They can work in the dark and are much harder to fool. While 2D infrared systems are only looking for heat, 3D systems also need depth information. The latter cannot of course be achieved without a reasonably accurate prosthesis figure.