CAMBRIDGE (CBS) – Theft is a huge problem for retailers, particularly in grocery stores where customers use self-checkout lanes. According to the Loss Prevention Foundation, stores that operate on a thin margin, have to sell dozens of items to make up for just one that was stolen.

In an effort to solve that problem, Malay Kundu tapped into his knowledge of visual technologies.

“I used to develop facial recognition systems to look for terrorists in airports,” he told WBZ-TV.

But instead of terrorists, Kundu is now looking for un-scanned items at grocery checkouts and other retail stores.  His Cambridge Company is called StopLift and the owner of Vicente’s grocery store in Brockton is a big fan.

checkout1 Cambridge Company Uses Artificial Intelligence System To Stop Shoplifters

(WBZ-TV)

“It has saved us a good amount of money,” said Brian Vincente.

Vicente’s store has a loss prevention room filled with dozens of cameras watching every checkout aisle and the entire store.  It would be impossible for the human eye to keep a close eye on all of them at once.

That’s where the computers come in.

checkout2 Cambridge Company Uses Artificial Intelligence System To Stop Shoplifters

Surveillance cameras at Vicente’s grocery store in Brockton. (WBZ-TV)

“Our software is analyzing the video and it is also looking at what’s actually being rung up on the checkout. It’s comparing those and it can tell when you are trying to give yourself a deal,” Kundu said.

The program can catch a number of different ways both self-checkout customers and cashiers cheat the system.

checkout4 Cambridge Company Uses Artificial Intelligence System To Stop Shoplifters

(Image credit: StopLift)

Kundu pointed to video of a woman ringing up bags of potato chips using the produce lookout system, paying just the price of loose carrots, which is considerably cheaper than the chips.

checkout5 Cambridge Company Uses Artificial Intelligence System To Stop Shoplifters

(Image credit: StopLift)

A second example shows a self-checkout customer pretending to scan several items and tossing them to the end of the belt to avoid the scale underneath used to detect the weight of items to make sure items are rung up properly.

Another self-checkout customer correctly scams several items and bags them. She then places bags on top of several items that were never removed from the cart and walks out of the stores.

According to Brian Vincente, the technology is not just about his own bottom line.

“The reason why we use StopLift is to make sure that there’s no theft and there’s no negligence and that way we can assure that our prices stay low for our customers and we’re giving them the best prices that we can,” he said.

checkout3 Cambridge Company Uses Artificial Intelligence System To Stop Shoplifters

(Image credit: StopLift)

According to Kundu, when stores spot an issue they will approach the customer and politely ask if they need help using the self-checkout system.

StopLift is used at hundreds of checkouts in stores across Massachusetts and thousands around the world.

Comments

Leave a Reply

Please log in using one of these methods to post your comment:

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s