Apple seeks to improve iOS security with image recognition

Apple is planning to improve the security of iOS on iPhones and iPads by using image recognition or response, according to a patent application

Apple is planning to improve the security of iPhones and iPads by using image recognition or response to unlock the devices, according to a patent application.

The US Patent and Trademark Office has published a patent application made by Apple in 2011 for a system that would require users to respond to a randomly chosen image to access devices.

The system is aimed at making Apple’s devices more secure and less attractive to thieves by replacing the current passcode with something that is not so easily compromised.

The proposed system will require users to either identify one of their contacts or it could be set up to display images that users associate with specific responses.

For example, according to the patent application, a picture of the Golden Gate Bridge could be associated by the device owner with the last four digits of a friend’s phone number in San Francisco.

So whenever the image appears as part of the authentication process, those four digits must be entered to access the device.

Apple is believed to be exploring other ways it could replace the passcode system of its iOS operating system, according to the Telegraph.

The paper notes that, in 2011, Apple acquired biometrics firm Authentec, indicating the possibility of fingerprint recognition.

Apple has also filed several parents related to facial recognition, whereby the iPhone or iPad would identify its owner from their facial features.

Facial recognition systems have yet to be perfected, however. Such technology has already been implemented in Android, but it is possible to unlock handsets using a photograph of the owner.

In 2009, researchers from Vietnam demonstrated that they could bypass facial recognition biometrics in this way at the Black Hat security convention in Washington DC.

Nguyen Minh Duc, head of the application security department at the Bach Khoa Internetwork Security Center at Hanoi University of Technology, demonstrated various ways of beating the facial recognition systems built into Lenovo, Toshiba and Asus laptops.

Read more on Endpoint security