POINT-AIM-SHOOT-WE KNOW YOUR NAME BECAUSE OF YOUR SKIN TYPE ETC

Online shopping and advertising already do it, take information based on the pages or products that a person had looked at and provide advertisements, or links to other products that may also interest that person.

In just a few years shops could use facial recognition technology to do the same.

A Perth professor is working on research that he hopes could play a role in creating this technology.

Associate Professor Ajmal Mian from the University of Western Australia first became interested in facial recognition technology when doing his PHD which he completed in 2006.

Since then he has continued to research how to use satellite technology to identify facial features that lie under the skin.

It is believed that a dot-sized part of a face may soon be all that is needed to identify a person.

Professor Mian said by incorporating numerous images of a person from different angles into a system, these could possibly be used to later identify that person by just a small section of their face.

He said while facial recognition technology was not new, being able to identify someone from just a small part of their face meant recognition could be done faster and easier.

“To be more useful it has to not be intrusive, so you don’t need to come in contact with it like fingerprinting and the ultimate is to do it without people noticing it’s happening, without them having to stop and look at a camera,” Professor Mian said.

“I am trying to dig out more accurate techniques and find different algorithms to be able to identify people more easily.”

He said a shop may use the technology to maintain a customer database.

“We know security cameras are there but if shops say you need to get fingerprinted, people are not going to want to do that,” Professor Mian said.

He said the technology may not necessarily associate people by their names.

“They may group you by different charts, they don’t necessarily have to attach a name to it, each time you come in they see what you buy, if customer A buys item such-and-such they are most likely to buy item such-and-such, like on Amazon,” he said.

Mr Mian said it was up to marketing staff as to how the information was used.

He said multi-spectral imaging can be used to measure light reflected off a face at hundreds of discrete wavelengths in the visible spectrum and beyond.

This meant that the technology being worked on would be able to recognise a person despite their different facial expressions.

Professor Mian said his research may also be able to detect people who have used cosmetic surgery to alter their looks.

He said he did not expect the technology to be expensive once created.

“Once the algorithm is developed it won’t be expensive, it is the research which is the expensive part, all you will need is a few cameras.”

“It’ll start up in shops that spend a lot of money on customer care and marketing and others will follow.”

He admitted that there would be some concerns about privacy.

“There’s always a concern about security and privacy and there’s always a trade off, it will be a discussion of topic forever,” Professor Mian said.

He said the kind of facial recognition technology he envisioned could be used in security and if used at airports could greatly improve the identification process at the immigration sections of airports.

Professor Mian was also looking into the possibility of applying it to psychology and also identifying whether people had certain syndromes.

Associate Professor Mian is the only West Australian to have won the Australasian Distinguished Dissertation Award from The Computing Research and Education Association of Australasia.

He has also won two prestigious national fellowships: the Australian Postdoctoral Fellowship and the Australian Research Fellowship.

Sourced & published by Henry Sapiecha

Leave a Reply