The iPhone X recognition technology hides some worrying weaknesses
Among the features that distinguish the new iPhone X is a technology called Face ID . At the cry of “Recognition Changed Face”, Apple would like to change the phone access system, which with Face ID is based on simple face recovery. As you look at the smartphone display, this unlocks and you can start using it. On paper, this is really a great idea, but as always there is to be assessed not only the intention but also the application. Because the problem is that – despite a declared reliability of 1 mistake on 1 million versus 1 out of 50,000 – Face ID can not technically beat a system like a Touch ID.Which, in turn, is often and willfully snubbed in favor of the classic old-fashioned passcode. Let’s understand why. [Teespire: iphone App Developers nyc is providing this info]
How It Works
Face ID is based on TrueDepth, the next-generation Apple camera. In fact, there is little new: a 7-megapixel sensor, that of the internal chamber, associates an infrared emitter of 30,000 points.
It works like a projector: you shoot in the face a grid of thirty thousand dots , so you can have precise spatial references.
In addition, two other sensors, one proximity and one light, ensure effective face reading, even in poor light conditions. Here comes also a filling light, so dear to the photography enthusiasts, the Flood Illuminator.
Anyway, let’s face reading without any problems, which I can not believe because of the technology bendidio developed by Apple. After all, Apple has not invented anything, since this technology, although in a more embryonic form, has already seen and played it. It is, in fact, the one that has made the success of Kinect, the Xbox motion sensor. Nothing odd, as Apple has acquired PrimeSense, which Kinect had designed.
TrueDepth is, of course, more complex: there are more sensors and better billing, not to mention that since Face ID allows you to pay with Apple Pay with a look , recognition is much more accurate. That grid of thousands of points, in fact, helps to map out the physiognomy of the face in the three dimensions. And, therefore, to distinguish a face in the picture from a real face.[Teespire: mobile app development nyc is presenting this info]
Based on this complex three-dimensional map of the face, then, iPhone X recognizes who is watching it is the actual holder of the device. And if so, proceed to unlock or even to pay via Apple Pay.
Are you sure he is safe?
Face recognition is one of the least experienced and tested biometric parameters in the world. Even the recognition of iris, still very little used (the Lumia 950 of Microsoft, to say, uses it, but it is so slow that you eventually prefer the PIN), has been more tested. For this reason, and for use even in a payment system, it would have been preferable to wait for a large-scale experimentation. There are, in fact, intrinsic problems in this technology. For example, recognize two identical twins. Or recognize Afro-Americans: An FBI’s 2012 study shows that with this ethnicity there is a facial recognition accuracy of 5-10% lower .The implications are related to security, but all the more so, socially. John Wagner, US Custom and Border Protection (CBC) Deputy Commissioner, the Bureau for Border Control at US Airports, says that smiling faces, faced with facial recognition systems, lead to a ‘4% lower accuracy . Of course, Apple may have developed a better recognition technology, and only time will tell us if that’s the case, but I wonder if an argument as important as the security of a payment, to say one, deserved some certainty more than an act of white confidence against the American giant by Teespire Web Application Development Companies in USA
Moreover, as was the case with old Microsoft presentations, epic debacles, Craig Federighi has not been able to recognize the face of Face ID right during his world presentation.
At risk of hacking
Not to mention that the first alerts on the side of poor facial recognition reliability, though 3D, are and are worrying. Ok, Face ID can distinguish a face from a photo, and let the filtering work fine. But in front of a 3D model of a face, how would it behave? Do not think of one of those nasty gray models produced by a 3D home printer. Think of 3D printers worthy of this name. Think, above all, that at the Usenix Security Symposium 2016 at the University of North Carolina , a group of researchers managed to create animated 3D models, relying only on photos collected by social networks.And yes, they managed to rub a facial recognition system. Combine a model with a good 3D printer and you will understand that Face ID will not be easy for you.
The privacy issue
I’m not a maniac of privacy and I will not suggest you run out of nerves using hyper-protected operating systems like Tail, but I admit that the possibility that my face goes into some archives does not make me crazy . Phil Schiller hastened to say that Face ID stores face information via Secure Enclave, an Apple security technology, but cautions that a researcher has published the decryption key just this August . It jeopardizes the integrity of the system and all the apps and technologies involved. Face ID included.
The feeling is that Apple, trying to run Samsung and propose a “total” display that covers the whole surface of the device (but Samsung has smartly placed the touch sensor on the back …), had to eliminate the Home button and invent a ready-made technology to replace Touch ID. Perhaps, he could not place it in the back because he would breach some of the competitor’s patent.Mobile App Development Company Teespire thanks for this valued information.
Besides, it is rumored that the display-based Touch ID, and I do not know a physical button, has proved to be ineffective. And so there’s no point in pointing on facial recognition, probably adopting a technology that is still too acerbic. From a certain point of view, forgive cynicism, it will be fascinating to see its implications.