What is Facial Recognition? Everything you need to know

Have you ever been to another country? If the answer is yes, then it’s most likely that your face was scanned at customs when entering that country. Have you ever wondered why they do that? Or how your Smartphone can unlock using your face instead of a password? All this is made possible using facial recognition technology (aka FRT).

What is Facial Recognition?

Instead of giving you one of those technical, hard-to-understand definitions, let me give you a simpler one. One person’s facial structure differs from the other. In other words, all of us have a unique facial structure, similar to the unique structure of a thumbprint. Facial recognition technology can analyze a person’s facial features, match it with information in a database, and identify who he/she is.

In ’18, the facial recognition market was at $3.97B, and by ’25, it is predicted to grow to $10.15B. People seem to be divided on this technology. While some are pushing for its increased use, others raise into question the security and accuracy.

How Does Facial Recognition Work?

Before I begin explaining how it works, you should know that this is just a general explanation of its working. There may be facial recognition technology systems out there that may function a little differently. With that said, let’s start:

Step 1: Face Detection

The process begins with a camera detecting and recognizing a face. Although the camera can detect a face in the crowd, it is best detected when looking directly at the camera. 

Step 2: Face Analysis

Once a face is detected, what happens next is that a photo of the face is captured and analyzed. As mentioned before, there are all sorts of facial recognition technologies out there. While some rely on 3D images, most of them rely on 2D images. Primarily because it’s easier and convenient to match a 2D photo with public photos or those in a database than a 3D one. Our faces are made up of nodal points (or distinguishable landmarks). There is a total of 80 of these. So, what facial recognition software does is to analyze the distinguishable landmarks such as the distance between your ears or the shape of your cheekbones.

Step 3: Converting An Image to Data

Okay, so once all the analysis of a face is completed, this analysis is then turned into a mathematical formula. All the features of the face become numbers in a code, which is referred to as faceprint. As mentioned above, all of us have a unique facial structure, which means all of us has our own unique faceprint.

Step 4: Finding a Match

The previous step (which is the faceprint) is then compared against a database of other faceprints. 

Here are some mind-boggling facts for you, The FBI (in case you don’t know, it stands for Federal Bureau of Investigation, it is the domestic intelligence and security service of the United States) has access to over 641 million photos—including driver’s licenses, passports, and mug shots—in a searchable form that can be used by facial recognition technology. Moreover, Facebook also has a large database of photos of people’s faces. Any photos that are tagged with a person’s name become part of the Facebook database.

The technology searches the database until it finds an exact match. Once it does, it returns with the match and any other attached information such as name, address, etc.

Where is Facial Recognition Used?

The applications of this technology are almost limitless. Let’s go through some of the ways it’s been used now:

Device Security

Almost everyone has a Smartphone today. If you do, too, then you must have come across mobile apps that use facial recognition to protect your data. No matter how secure a password is, it can still be cracked. A reason why lots of apps have turned to facial recognition. These apps use the phone camera to scan a person’s face and only provide access to the Smartphone or private data if it’s the correct user.

Identifying Genetic Disorders

Face2Gene is a healthcare app that utilizes facial recognition technology to identify rare genetic disorders patients from just photos. Created by FDNA (a US biotech company), this groundbreaking technology will help prevent patients from enduring lengthy waits and making repeated clinic visits before they are finally diagnosed – giving them a better chance of receiving the right treatment early on. This app should be used solely by healthcare professionals and should not be used without proper medical training.

Shoplifting

More and more stores are getting facial recognition systems that mark those individuals who have shoplifted before as a threat. So, if a shoplifter enters a store, this system will identify him and will notify the store owner of their past discrepancies, even if that shoplifter has never entered that specific store before. This system is not accurate, making it very dangerous because if an innocent person gets mislabeled as a thief, it could impact their whole life.

Buying Alcohol

There are bars and grocery stores in the UK that use this technology to identify if its customers are old enough to have a pint. The grocery stores allow customers to use the self-checkout for alcohol without needing an extra employee standing by to check IDs. Those customers who are identified as younger than 25 need to have their IDs checked. 

App That Makes You Old

Recently, my social media feed was flooded with the gray-haired faces of seniors. I’m pretty sure yours one was too. The reason behind it was FaceApp, an age filter that uses facial recognition to age your face. Maybe you used it too. Although it’s quite an interesting app, unfortunately, there are concerns that the data it collects from users is not secure, so use them at your own risk.

What Are the Concerns Of Facial Recognition?

As mentioned above, facial recognition has made its way into lots of different industries. Still, there are many people out there who are hesitant to accept this technology, primarily because of concerns about errors in recognition, privacy, and misuse of data. 

Errors in Identification

There are times when facial recognition fails to accurately match faceprints with the database, primarily because of poor images. It becomes harder to analyze the person’s nodal points accurately. This creates an error in the faceprint, making it impossible to match the database’s right data. 

Privacy

Many privacy issues come with facial recognition, which has got many people concerned. The technology can track you, which makes people uneasy. Moreover, large data breaches have become a common phenomenon these days, and the personal info that facial recognition software collects is too not immune to these data breaches.

Misuse of Data

Not only do private entities have access to the data collected by their facial recognition software, but lots of facial recognition databases are public. This means that anyone can find you on the database and track you down, even those with malicious intent.

Thank you for reading!