CIO Insight has a story on how Retailers are using cameras and pattern recognition software to fight back against shoplifters who try to hide items in the compartments underneath grocery carts.

Here's how the system is designed to work: When the shopping cart enters the checkout lane, it passes by a small lane-mounted camera that records images of the products beneath the cart and compares them with a database of known products that are considered likely to be stored underneath the cart, Sakaguchi said.
Using visual pattern recognition—which considers the colors, shapes and images on the product but not its barcode or RFID tag—the system tries to identify the product and then automatically puts that product into the point-of-sale system. If it works, the item automatically appears on the display just as if its barcode had been scanned by hand, Sakaguchi said.