DA3. Biometric Authentication¶
Statement¶
We are surrounded by digital devices. Look around yourself and identify a device with a biometric authentication method, and:
- Provide a detailed description of the device, the method of authentication, and how you use it. Share your experience of using this biometric authentication.
- Identify and discuss the technical details of this method, the feature vector, the authentication process, and the accuracy of the method used. Determine if this authentication method could be bypassed, and/or if there are any other issues related to biometric authentication.
Solution¶
I use an iPhone device as my primary phone; I used to unlock using a pattern that you have to draw on the screen. However, I switched to using FaceID, which is a biometric authentication method. FaceID is a facial recognition system that uses a TrueDepth camera system to scan the user’s face and compare it to the stored face data (Apple, 2023).
I use FaceID to unlock my phone quickly; it is much faster than typing in a PIN or drawing a pattern. It works in all cases, during day or night; when I’m wearing a mask or sunglasses. The phone has a fallback to using a PIN which is used when I hand the phone to some family member and I’m not around. Sometimes, if the front camera is covered or has some dirt on it, FaceID fails to authenticate, and I have to enter the PIN.
The technical details of FaceID involve TrueDepth projecting and analyzing thousands of invisible dots to create a precise face map, along with capturing an infrared image of the face. Another neural engine takes this data and transforms it into a mathematical representation that is comparable. FaceID increases security by allowing only 5 failed attempts, using an attention detector that requires the user to be looking at the phone to unlock it, and encrypting the face data and storing it in the Secure Enclave (special hardware included in the device) (Apple, 2023).
Advantages of using facial recognition(AWS, n.d.):
- It is more efficient and faster compared to other biometric methods, like fingerprint or iris scanning.
- It is more accurate than using traditional methods, like phone numbers, or IP addresses.
- It is easy to use and integrate with the existing echo system of smartphones and other devices.
Disadvantages and issues related to facial recognition:
- Its accuracy is near-perfect in good circumstances, but it may be affected by the environment like the light, the angle of the face, etc. (AWS, n.d.).
- Its accuracy may get reduced with age, as the face changes over time (Stouffer, 2023).
- Privacy concerns over data breach, however, in the case of Apple FaceID, they claim that data is stored only in the device and never shared, also data is encrypted in special hardware (Apple, 2023).
To conclude, facial recognition is a good biometric authentication method, and its market is expected to grow to up to $16.74 billion dollars by 20230 (Stouffer, 2023); but it is not perfect, and it can be bypassed in some cases for example, using a 3D printed face mask (Stouffer, 2023) or by a twin that has a similar face.
References¶
- Apple. (2023). About Face ID advanced technology. Apple. https://support.apple.com/en-gb/102381
- AWS. (n.d). What is facial recognition? Amazon Web Services. https://aws.amazon.com/what-is/facial-recognition/
- Stouffer C. (2023). What is facial recognition and how does it work? https://us.norton.com/blog/iot/how-facial-recognition-software-works