Innovation: camera facial recognition technology

The advent of artificial intelligence and facial recognition is bringing cameras into the front line of foodservice. Jim Banks looks at these new applications

Cameras are everywhere – in the street, our homes, our phones. Estimates of the number of photos taken since the advent of the smartphone vary, but it is no surprise to find that the generally accepted number is more than one trillion every year – and rising.

People alive now are the most photographed in human history. We have become accustomed to being on camera, whether we are taking selfies or being monitored by security cameras.

Surveillance for security purposes is nothing new and is a familiar function in shops, hotels, restaurants and even some homes. The deployment of cameras, however, is starting to go far beyond passive monitoring.

“The use of cameras has changed over the years,” says FCSI Associate Jay Bandy, president of Goliath Consulting Group, based in Georgia, US. “Initially, they were used at the security level to keep track of what people were doing. Then they were linked to the POS systems, watching servers ring up orders, and as part of anti-theft measures.

“Cameras then moved outside the building, particularly in the QSR market, to look at the number of cars queuing for the drive-thru or tracking licence plates on cars to identify customers,” he says. “Now, they are starting to come back inside. With facial recognition capability, it is possible to instantly recognize a customer and suggest orders, upsell and gather more data about that individual.”

The new applications of cameras in foodservice derive from the rising tide of technological innovation – not so much in the cameras themselves, but in the software that analyzes the images they capture. Artificial intelligence (AI) is utilizing its ability to crunch vast amounts of data.

“Cameras are clearly linked to AI,” says Bandy. “As wages are going up – the minimum wage has risen here in the US – restaurants need to stay profitable and the use of technology, including AI will become more palatable as a way to improve their P&L. We have seen it in voice recognition technology, which already ties into POS and ordering systems pretty seamlessly.”

Behind the scenes

With AI, cameras are becoming the eyes of smart systems that are increasingly focused on driving productivity and efficiency. “The baseline for using cameras is security and monitoring, but we have much more sophisticated systems now,” says Juan Martinez FCSI, principal at industrial engineering consultancy Profitality in Florida, US. “We are hearing that it is possible to measure labor efficiency before and after changes are made to restaurant design and workflow.

“Cameras can track bodies and shadows, so they can sense when people are active or idle,” he adds. “It is not widely used, but it is a big conversation that is going on in the industry about cameras and AI, though we are not yet sure how it will be used.”

Facial recognition capability is, for now, the high point of the marriage between cameras and AI. The ability to pick out an individual with no other information than a picture of their face has been seen in a thousand spy films, but in foodservice it offers very powerful and much less sinister opportunities.

Cameras can already identify a particular vehicle by its licence plate, which helps to track orders and, potentially, personalize service. Facial recognition capability takes the same principle even further. Yet, at a time when many people are wearing masks to prevent the spread of Covid-19, this may not yet prove to be the killer app.

A technique similar to facial recognition is, however, being implemented to check that people are wearing masks. Dragontail Systems has launched a Covid-19 enhancement for its QT AI camera, which monitors food preparation processes.

Dragontail, which optimizes restaurant order and delivery process for global brands such as Domino’s and Pizza Hut, has updated its AI-based computervision quality management system to detect cleanliness and sanitation. The existing QT AI camera system monitors the quality and accuracy of ingredients, as well as the temperature of the food. The latest iteration now detects the presence of gloves and masks, as well as keeping track of how often equipment and workspaces are cleaned.

With Covid, less contact and better food safety are top priorities, and Martinez points out that such a system could be used to help with adherence to sanitisation protocols. Logging when someone has washed their hands and sending a reminder if it has not been done within a specific timeframe is just one potential application. It is his firm belief, however, that camera technology will have the biggest impact in the area of efficiency.

“In Florida, the minimum wage is rising from $9 per hour to $15 over the next few years, so the cost of labor is rising,” he remarks. “If restaurants can only afford fewer employees, then they may turn to cameras and AI to increase sales while cutting back on labor.

“The technology could help boost efficiency both internally and externally,” he adds. “It could help to reduce labor costs and, at the same time, improve the speed and quality of service, order accuracy and throughput. There is a need for that kind of efficiency.”

Consumers embrace the camera lens

Domino’s, Chipotle’s and McDonald’s have been quick to move on AI, voice and camera technology. Although not all have implemented AI-enabled cameras, it is clear that big brands in the QSR segment will lead the way in adopting such technology.

McDonald’s is acquiring Apprente, a company that uses artificial intelligence to understand speech, and could implement its technology in drive-thrus, self-order kiosks and its mobile app. Big chains are also using licence plate recognition for drive-thru or kerbside delivery, and it is a small step from there to recognising a customer’s face to personalize orders.

“This is an area where AI could be used to suggest an order and we will certainly see this use of camera technology more in the future,” says Martinez. “Facial recognition is a technology I can see being used more and more.”

PopID cameras have already become popular as a contactless means of payment in some parts of the world. Once a user signs up and take a selfie, the camera in a kiosk can recognize their face in its cloud database and enable ordering, payment or entry through a security gate without a password or a debit card. “Older generations and some segments of the younger generations don’t trust face recognition technology,” notes Bandy.

“But with payment applications like PopID it is becoming more familiar.” “With PopID, the camera identifies you and charges your card, so all you have to do is look into the camera and pick up your food,” he adds. “People are very enthusiastic about the ease of use and it could not fit better with the current Covid-19 environment. Everything is contactless until you pick up your order. You don’t even have to hand over your card.”

Beware big brother

Surveillance and facial recognition technology do carry authoritarian overtones for some customers. There are moral and ethical dilemmas that arise, not least around the right to privacy. In the US, there have been facial recognition protests and a handful of cities have banned use of facial recognition technology, at least by the government. “Here in Florida, they introduced cameras on traffic lights, so if you ran a red light you got sent a fine,” notes Martinez.

“People complained and the cameras were outlawed. But facial recognition is very important and if you’re not doing anything wrong, why would you object to it? Many people already volunteer so much information about themselves on social media.”

While some are vocal in their opposition, the truth is that most of us have become so accustomed to cameras that we no longer notice them. “Over the years, people working or dining in restaurants have become so used to cameras being there, so they ignore them. The use of cameras and AI will continue to expand and will be seamlessly integrated into the restaurant infrastructure. Domino’s and Chipotle are leading the way, but it will steadily filter down to smaller chains,” says Bandy.

“You want to be leading edge, not bleeding edge,” says Martinez. “You want to be in the first wave, but not the first to make a move. There is a cost to the technology, not just the upfront cost, but also repair and maintenance. If it were cheap, it would be used everywhere already. One thing is for sure, you can’t ignore the progress of technology.”

Jim Banks

More Relevant

View More