Woody Bledsoe, a computer scientist, developed a method in the 1960s to classify faces using gridlines. In this method, which is considered the beginning of the face recognition system, it was a difficult task to extract the coordinates of the facial features and enter them into the computer. Towards the end of the century, a new high-accuracy technology was developed at the University of Bochum, Germany. Banks and airport operators first preferred this technology. In the past, due to the low resolution of the camera, the desired image could not be captured and it was difficult to match. However, today, thanks to high-resolution cameras, face identification can be done easily.

What is Face Recognition Technology?

The human face has different characteristic features in each individual. Each individual's eyes, facial line and nose structure differ from each other. Face recognition is a computer-based security system that detects any person's face thanks to these differences and measures the overall facial structure. In short, face recognition is the detection technology, within the scope of biometric application, the verification of one's unique face data by matching on the database side and analyzing the contours of the person's face. Its usage area is generally on security.

How Do Facial Recognition Systems Work?

The human face consists of about 80 nodes. Face recognition technology, distance between eyes, width of nose, depth of eye sockets, shape of cheekbones, length of jawline etc. is measured as. Experts usually focus on the golden triangle between the lips and temples. This part remains the same in case of aging, weight loss or gain, hair and beard growth, even wearing glasses. Basically, the system can be examined in 4 areas;

⦁ Face identification : During recording, the system takes a physical or behavioral sample.
⦁ Parsing : Unique data is extracted from the sample and the template is created.
⦁ Comparison : The template is compared with the new sample.
⦁ Matching : The system decides whether it matches the features from the newly arrived sample.

Here, after the face is detected first, the background is smoothed. The size and position of the face are normalized to values ​​as previously detected. The image is standardized as a pose so that the accuracy of the face can be compared with the information received in the database. In order to make this standardization, it is important that some signs (circular shapes) on the face can be defined correctly. Mathematical representation of image data is provided by algorithms.

Where are Face Recognition Systems Used?

The facial recognition system can be used as authentication in election records, in the banking sector, in many public and private sectors such as electronic commerce, national identities, passports, employee IDs.

It is among the methods that can be used to increase the level of mobile phone and computer security.

Thanks to facial recognition systems installed in vulnerable areas in markets and large stores, suspects can be identified quickly, providing the opportunity to catch criminals by recognizing their faces.

Face recognition systems can also be used in marketing and campaign activities.

Thanks to the systems implemented by international airport companies, there are scenarios where the flight tickets of the passengers are scanned and recorded through the face recognition system.

Why Should We Use Face Recognition System?

⦁ It is secure and fast in authentication processes. It eliminates problems such as card theft, corruption, forgetting password or sharing all of these.
⦁ Frauds can be prevented and criminals can be detected with facial recognition.
⦁ Easy to use. Accuracy rate is high.
⦁ It is very important in terms of hygiene because there is no contact.

Conclusion

As we said at the beginning of our article, the facial structures of all individuals are unique. Therefore, the face shape of another person cannot be copied and used. This is an important proof of how reliable the face recognition system is.

On the other hand, traditional attendance and monitoring systems require physical intervention. An infected person touching a machine or card with an unclean hand increases the risk of infecting others. The use of facial recognition systems eliminates this risk, as there is no physical contact.

According to the researches, it is predicted that its usage will increase by 70% as of the coming years, thanks to its low costs, easy use and hygienic use.