One of the fastest developing security technologies is face-detection-based software. Under the umbrella of face detection, we find face recognition as well as face verification and although these technologies’ names are often used interchangeably, they are all distinctly different in their capabilities and uses. All three are important tools in ensuring security and fraud prevention both in-person and digitally.
What is face detection?
Face detection is the foundation for face recognition and face verification. This is something many of us are familiar with in our smartphone cameras’ autofocus features. Face detection is simply the system’s ability to detect if a human face is present when it is presented with a myriad of different items at once. In the case of your smartphone camera, the system scans the picture frame looking through background items such as trees or cars, etc. as well as what’s in the foreground such as your face, arms, and other body parts.
How it works
Face detection works by analyzing the image with algorithms designed to pick out features of a human face such as eyes, mouth, and nostrils. As mentioned, smartphone cameras typically have this ability, but there are other uses as well. At IDScan.net, we offer a temperature reader that uses facial detection to automatically recognizes when a visitor stands in front of the device, allowing the device to take the visitor’s temperature completely touch-free. Facial detection software does not have the ability to recognize a face as distinct from other faces, but it is the first step in recognizing a face as belonging to a particular individual. In the case of this temperature reader, facial recognition is used once a face is detected to keep a record of who the guest is, what their temperature was, and what date and time they visited. But what is the next step forward in this technology, known as facial recognition?
What is face recognition?
Face recognition is when a system scans for human faces and takes a digital image of the face as it did in the face detection process. It then compares the image to a database of other faces recorded over time to try and find a match.
How it works
How exactly does the technology know when it has found a match? Face recognition analyzes each face it comes across and determines each one’s particular facial feature geometry; that is, the distance between the eyes, between the brow and chin, etc. Each human face has a different layout so the system can compare every face it sees to its database of recorded faces to find a face recognition match. The system can build its database using ID card photos, images taken from security camera footage (either live or previously recorded), or from another photo.
Face recognition can be used live to enforce security in a building or at a venue by scanning groups of people to looking for familiar, unwanted faces. For example, it can help to identify known shoplifters it’s seen before on camera or from an image shared by other retail locations. But it can also help to identify repeat customers and improve customer experience. IDScan.net offers an innovative live video stream face recognition system that captures the image of a person’s face using it to recognize them in a live video stream. Our software can alert you the next time that person is recognized which can help businesses welcome frequent visitors or VIPs and help keep previously banned patrons from entering the premises.
What is face verification?
Face verification is a technology that uses face recognition authentication to make sure someone is who they say they are when they present their identification documents digitally. For goods being sold on a digital platform that require identification from the customer, making sure that the person pictured on an ID or passport is indeed the customer is vital. But how can we know for sure that this is the case? That’s where face verification comes in.
How it works
Face verification is used to map the face in the selfie, as well as the face in the ID photo, and compare the two images to calculate a confidence percentage in the match. With IDScan.net’s solution, your company can determine for itself what face verification match percentage is acceptable for your standards and our team at IDScan.net is here to help you make that decision. It’s also easy for both you and your customer to complete the face verification process. There are three easy steps for your customer: take a picture of the front and back of the ID card with their smartphone and take a selfie for face verification purposes. This information is instantly in the system with no input needed from you, the business owner. A face verification confidence percentage is then generated to automatically accept or reject the results, or to be reviewed.
This method of face verification for the purpose of verifying identity is incredibly useful for anyone selling goods online or for onboarding customers applying for credit or loans. Face verification helps to keep your company Know Your Customer (KYC) compliant and IDScan.net’s mobile identity validation solution not only verifies that the customer is who they say they are, but it also checks to ensure that the ID itself is legitimate by assessing the information stored in the ID’s barcode. This helps to prevent fraud that is an ever-increasing problem for businesses – especially those that operate digitally.
Although face detection, face recognition, and face verification all have to do with complex algorithms scanning faces, the terms should not be used interchangeably. They all have their different uses and each serves its own important function. Face detection checks to see that a human face is present in an image, face recognition determines if the face is familiar based on a photo database, and face verification checks to see that two images match each other, ensuring that the customer is who they say they are.
No. Although the technologies are similar, each term refers to a different technology.
Face detection checks to see that a human face is present in an image, face recognition determines if the face is familiar based on a photo database, and face verification checks to see that two images match each other, ensuring that the customer is who they say they are.