Foxconn could soon use AI to help identify faulty iPhone parts

By

Foxconn Wisconsin
New technology was designed by former Google AI expert.
Photo: Foxconn

Apple manufacturer Foxconn is planning to adopt artificial intelligence image recognition systems for quality control in its factories, a new report claims.

The tech could help identify faulty circuit boards or other components, thereby improving Foxconn’s efficiency when it comes to assembling devices. The AI system was developed by Andrew Ng, a machine learning expert who has previously headed up major projects for both Google and Baidu.

At a press briefing in San Francisco, Ng’s Landing.ai technology was demonstrated spotting a defect in a circuit board photographed by a camera. In contrast to regular computer vision systems which required many thousands of sample images to train them to recognize a particular object, Landing.ai can do this with just five training images. That makes it far more suitable for a factory line producing multiple parts which may change on a semi-regular basis.

In his most notable previous work, Ng helped develop the Google Brain project in 2011, which was able to learn to identify objects such as cats by watching videos on YouTube.

It’s not yet clear whether Foxconn’s new camera tech will be used or Apple assembly, although it would make no sense for it not to be. Over the past several years, Foxconn has been active in pushing for more and more automation in its factories. From 2010, Foxconn has been talking about replacing its human workers with robots. It began testing this technology iPhone-building bots back in December 2012, and by last year was reportedly using 40,000 of these robots at its factories in China for handling manufacturing jobs.

Source: Reuters

Newsletters

Daily round-ups or a weekly refresher, straight from Cult of Mac to your inbox.

  • The Weekender

    The week's best Apple news, reviews and how-tos from Cult of Mac, every Saturday morning. Our readers say: "Thank you guys for always posting cool stuff" -- Vaughn Nevins. "Very informative" -- Kenly Xavier.