Facial recognition used to be a staple for science fiction novels, TV, and movies.
Technology has leapfrogged to the point where all the spectacular ideas of sci-fi are now as ubiquitous as a toothbrush. Today, you’re apt to find facial recognition everywhere – from unlocking your smartphone to posting photos on Facebook. How can facial recognition help the security of businesses? Before answering that, let’s take a quick look at what facial recognition is and how it came to be.
The Evolution of Facial Recognition
Facial recognition technology has been around for quite a while now.
The Viola-Jones Object Detection Framework of 2001 was the first real stab at detecting faces with the use of algorithms. The framework’s initial design was for identifying objects within images. As the use of the system grew, engineers applied the algorithm to face detection with great success.
The framework’s facial detection prowess became popular because it was fast. The only issue was that training the machine to detect a myriad of facial features was very slow. In 2001 – 2004, the success rate of the algorithm running on a regular desktop computer was already at an astonishing 90%.
The real breakthrough in facial recognition software came in 2010. The Convolutional Neural Networks secured the top spot as the best method for facial recognition. More raw processing power and cloud computing contributed to the rise in detection rates. Computers were now beating humans in facial recognition.
How Facial Recognition Works
Facial recognition involves a complex, multi-step process involving specialized sub-systems. These are:
Detection and Tracking
The first step is all about processing. This stage handles tracking and identifying faces on images or video files. This step will tell the system if there’s a face that needs handling. This stage is also responsible for tracking facial expressions and unique facial features.
The second step handles wherein the image or video the face lines up. This stage also seeks out the contours and features of a face to determine whether it’s facing front or sideways.
The third step of the process handles the extraction of all the individual features of the face. The distinctive features are the eyes, nose, lips, chin, ears, and other identifying marks. The input collected by the algorithm at this point would be enough to tell faces apart by their unique features.
Feature Matching and Classification
The final stage handles the input received from facial extraction. The algorithm matches all information against a database to make a positive ID of a person. The software will know whether a face is part of a database or not.
Facial Recognition for Consumer Engagement
For consumer engagement, facial recognition allows more personalized and interactive content.
There’s a scene in the Tom Cruise movie Minority Report where he was on a floor where the ads changed as he passed by. We’re pretty close to achieving this level of personalization, although companies shouldn’t focus on ads.
Consumers are wary enough of pushed advertisements as it is, so employing this approach can backfire. Plus, there’s the issue of privacy. The better use of facial detection is by creating more engaging experiences. Companies already know a great deal about users.
Pushing an ad that doesn’t feel like an ad will serve the purpose of engagement without the push-back. A choose your adventure type of game/ad campaign works wonders.
Facial recognition isn’t perfect, and some criminals have the means to fool the system. However, as a business person, you are responsible for protecting customers from cyber-attacks and identity theft. Therefore, exploring all methods to protect your clients should be your top priority, one of these methods are criminal records which can be the best solution for fighting against crimes.
How Businesses can Use Facial Recognition
Facial recognition can have a positive impact on businesses when it comes to business security and consumer engagement.
Businesses attract criminals like moths to a flame. Whether you own a brick-and-mortar store or do all your business online, the threats are real. Facial recognition can help secure your interests and protect your customers.
For retail stores, facial recognition technology can thwart would-be shoplifters from stealing. CCTV cameras with face detection tech will actively scan the retail floor. The algorithm will seek out known thieves on the database, and alert security once spotted.
Preventive measures for retail security are good, but what about shoplifters not in the database? As the CCTV scans the floor, it records all data. If an undocumented thief does make it out with stolen goods, the system stores the incident. The face of the shoplifter is then added to the database and forward to the police.
Security for e-Commerce
Facial recognition provides an extra layer of protection for e-Commerce businesses. Online stores are prime targets for hackers and fraudsters. Hackers can steal credit cards and other client data they can use for Identity theft. To counter this, businesses can us a “Smile to Pay” feature like the one Ant Financial began using in 2018.
Paying with your face is simple enough. In the case of Ant Financial and Alibaba, users already had the Alipay app on their phones. The users only needed to enable facial recognition scanning on the app for the technology to work. When enabled, users simply went to merchants that accept Smile to Pay for seamless payments.
The system used in stores have a 3D camera installed in the POS system. The camera scans the customer’s face to verify their identity. Deductions are then applied to their Alipay account.