قالب وردپرس درنا توس
Home / Tips and Tricks / How does face recognition work?

How does face recognition work?



  The face of a woman with a grid. This grid is used to identify her face.
Stanislaw Mikulski / Shutterstock

Most people are comfortable with face recognition for use in Instagram filters and face recognition. But this relatively new technology is a bit scary. Her face is like a fingerprint and it's hard to understand how facial recognition works.

As with any new technology, facial recognition also has disadvantages. These drawbacks are becoming increasingly apparent as military, police, advertisers, and deepfake developers seek new ways to use facial recognition software.

More than ever, it is crucial for people to understand how facial recognition works. It is also important to know what facial recognition limitations are and how they will evolve in the future.

Facial recognition is surprisingly simple

Before you go into the many different facial recognition media, it is important to understand how the face recognition process works. Detection works. Here are three applications for face recognition software and a simple explanation for recognizing or identifying faces:

  • Basic Face Detection : In Animoji and Instagram filters, your phone camera searches for the defining features of a face, especially a pair of eyes, one Nose and a mouth. Then algorithms are used to capture a face and determine which direction it is looking at, whether its mouth is open, etc. It is worth noting that this is not face recognition, but only software that looks for faces.
  • Face ID and similar programs : Setting up facial recognition (or similar programs) on your phone will take a picture of your face and measure the distance between your facial features. Every time you unlock your phone, it "looks" through the camera to measure and confirm your identity.
  • Identifying a stranger : When an organization wants to identify a face for security, advertising, or advertising For control purposes, algorithms are used to compare that face with an extensive database of faces. This process is almost identical to the Apple Face ID, but on a larger scale. Theoretically, any database could be used (ID cards, Facebook profiles), but a database of clear, previously identified photos is ideal.

Okay, let's go into detail. Since the "basic facial recognition" used for Instagram filters is such a simple and harmless process, we focus entirely on facial recognition and the many different technologies with which a face can be identified.

Most face detection relies on 2D images

As you might expect, most face recognition programs rely solely on 2D images. However, this is not done because the 2D face imaging is very accurate. This is for the sake of usability. The vast majority of cameras take photos without depth, and public photos that can be used for face recognition databases (such as Facebook profile pictures) are all in 2D.

  A man using face recognition technology to identify a motif from a database.
Zapp2Photo / Shutterstock

Why is 2D facial imaging not very accurate? Well, because a flat image of your face has no identifying features like depth. With a flat image, a computer can measure the interpupillary distance and the width of your mouth, among other things. However, it can not detect the length of your nose or the protrusion of your forehead.

In addition, 2D facial imaging relies on the visible light spectrum. This means that 2D facial imaging will not work in the dark and may be unreliable in restless or shady lighting conditions.

It is clear that some of these shortcomings are resolved through the use of 3D facial imaging. But how is that possible Do you need special equipment to see a face in 3D?

IR Cameras Give Your Identity More Depth

While some facial recognition applications are based solely on 2D images, it is not uncommon for facial recognition to rely on 3D imaging as well. In fact, your facial recognition experience is probably a pinch of 3D.

This is achieved by a technique called lidar, which is similar to sonar. Essentially, facial scanning devices like your iPhone blow up a harmless IR matrix on your face. This matrix (a wall of lasers) is then reflected off your face and picked up by an IR camera (or ToF camera) on your phone.

<img class = "wp-image-428013 size-full" data-pagespeed-lazy-src = "https://www.howtogeek.com/wp-content/uploads/2019/07/3.png. pagespeed.ce.on6tzJIVDv.png "alt =" A woman using facial recognition or similar IR-based facial recognition technology. [19659023] Prostock Studio / Shutterstock

Where does 3D magic happen? The IR camera on your phone Measures how long it takes for every bit of IR light to be reflected off your face and return to the phone Of course, the light reflected from your nose has a shorter path than the light reflected from your ears and the IR camera uses that information To create a unique depth map of your face, in combination with basic 2D imaging, 3D imaging can greatly enhance the accuracy of the face recognition software.

Lidar imaging is a strange concept that can make it difficult to understand Head around. If it hi Try to imagine that the IR network of your phone (or face recognition device) is a whiteboard toy. Like a pinboard toy, your face leaves a deepening in the IR net where your nose is noticeably deeper than, for example, your eyes.

Thermal Images Enable Face Detection at Night

One of the disadvantages of 2D face detection is that it relies on the visible spectrum of light. For laymen, the basic face recognition does not work in the dark. However, this can be bypassed using a thermal imager (yes, as in Tom Clancy).

"Wait a minute," you might say, "Does the thermal imager not rely on IR light?" Yes. However, thermal imagers do not emit rays of IR light. They simply detect the IR light emanating from objects. Warm objects emit a ton of IR light while cold objects emit a negligible amount of IR light. Expensive thermal imaging cameras can even detect minor temperature differences on a surface, making it ideal for face detection.

  Three photos. The first comes from the visible light spectrum, the second is a still image and the third is a composite thermal image.
A visible light spectrum, a thermal image and a composite thermal image. Polaris Sensor Technologies Inc

There are a handful of different ways to identify a face with thermal images. All of these techniques are incredibly complicated, but have some basic commonalities. We will therefore try to keep things simple with a list:

  • Multiple Photos Are Needed : A thermal imaging camera takes multiple pictures of a subject's face. Each photo focuses on a different spectrum of IR light (long, short and medium waves). Typically, the longwave spectrum provides most of the facial details.
  • Blood vessel maps are useful : These IR images can also be used to extract the formation of blood vessels in the face of a person. It's scary, but blood vessel maps can be used like unique facial fingerprints. They can also be used to determine the distance between the facial organs (if the typical thermal imaging provides poor images) or to identify bruises and scars.
  • The subject can be identified : A composite image (or record) is created with multiple IR images. This composite image can then be compared to a face database to identify the subject.

Of course, thermal facial recognition is usually used by the military, and at Khols you will not find anything that will come with your next cell phone. In addition, daytime thermal imaging (or generally good lighting environments) does not work well, making it suitable for a few non-military applications.

Limitations of Face Detection

We spent a lot of time talking about the flaws of facial recognition. As we have seen from IR and thermal images, some of these limitations can be overcome. However, there are still some issues that are not yet resolved:

  • Disability : As you might expect, sunglasses and other accessories can trigger the face recognition software.
  • Posen : Face recognition works best with a neutral, forward-looking image. Tilting or twisting the head can make facial recognition difficult even for IR-based recognition software. A smile, bloated cheeks, or another pose can also change the way a computer measures your face.
  • Light : All forms of facial recognition are based on light, regardless of whether it is visible or IR light. As a result, strange lighting conditions can reduce the accuracy of facial recognition. This may change as scientists are currently developing a sonar-based face recognition technology.
  • The Database : Without a good database, face recognition can not work. In this sense, it is impossible to identify a face that has not been correctly identified in the past.
  • Data Processing : Depending on the size and format of a database, computer identification may take a while to get it right. In some situations, such as policing, data-processing limitations restrict the use of facial recognition for everyday use (which is probably a good thing).

From now on, the best way to get around these limitations is to use other forms of identification in conjunction with face recognition. Your phone prompts you to enter a password or fingerprint if your face can not be identified, and the Chinese government uses ID cards and tracking technology to close down the error rate on their face recognition network.

In the future, scientists will surely find a way to circumvent these problems. They can use sonar technology with LIDAR to create 3D face maps in any environment, and they can find ways to process face data (and identify strangers) in an incredibly short time. In any case, this technology has a great potential for abuse, which is why it pays to keep up to date.

Sources: The University of Rijeka, The Electronic Frontier Foundation


Source link