What Is Facial Recognition?
Lately, more facial recognition systems have been popping up in all kinds of places. The courts are using facial recognition software to identify people at crime scenes, police departments use it to track people, and there's even a video security camera system that uses it to identify bad guys. The potential for its use in other areas is almost limitless.
But what exactly is facial recognition technology? Is it really more powerful than the old systems, or does it just sound like something from an Arnold Schwarzenegger movie?
Facial recognition and artificial intelligence
In reality, it's not much different from how artificial intelligence works. In fact, AI can work on quite a basic level: your computer is using a sort of database to match a particular face against photos stored in it. For example, when you're looking at your photo album, you might see the name of the person you're looking for, then a thumbnail view of their face, and finally a closer view.
The system's main goal is to weed out the pictures of people who aren't your ancestors. It looks at the facial features and hairstyle of these images, then matches them up with the person whose photo you're looking at.
When artificial intelligence tries to do the same thing, it uses the same sort of database to scan faces. The goal is to match them up, so it can identify a face based on the data it has. Some of the systems are better at this than others, but the end result is that they're almost always able to identify people.
Facial recognition, as it turns out, is a pretty good way to do this. When it tries to match a face against photos and faces, it can either pull up a list of results, or it can try to match it with a template that's already been built.
Match by multiple elements
These templates have things like "name," "age," and "age range." The system might try to find a match by checking if the name and age match up, and if so, it could look at the person's name and age range. The people who are older will generally have more distinguished names, which means that it can get more accurate results.
What's great about this system is that it can learn from its mistakes. If it finds that a certain set of pictures doesn't match up, it won't be as likely to use it again. And if it gets better at matching, then it can quickly become more intelligent than anyone else.
This is the one weakness of the high-tech facial recognition systems that are being used in courts now. The problem is that they've only been able to learn from a small number of examples. It's a lot harder to know that a mug shot of a convicted felon is actually a convicted felon than it is to know that a mug shot of a visitor isn't a visitor, and so on.
One good thing is that we can hope to make a person less recognizable. Facial recognition technology, as I said, is pretty good at learning about facial features, but it's far from perfect, and it's not necessarily always possible to get a real-life photograph of someone.
Mass recognition technology
Fortunately, facial recognition technology can be used to identify people in large collections of photos, rather than just one or two. The larger the collection, the better the system is likely to get. And when it's working well, it can identify someone in as little as a couple of seconds.
Still, it's important to understand that facial recognition is a way of collecting data about a person using artificial intelligence. It works with similar principles as to how artificial intelligence systems work to collect data. So as a person becomes more aware of it, they can use it wisely, even if they don't understand the concepts behind it.