Biometric identification in integrated security systems.

#Biometric identification

Tatarchenko Nikolay Valentinovich
Timoshenko Svetlana Vyacheslavovna

BIOMETRIC IDENTIFICATION IN INTEGRATED SECURITY SYSTEMS  

Everyone knows the scenes from science fiction films: the hero approaches the door and the door opens, recognizing him. This is one of the clear demonstrations of the convenience and reliability of using biometric technologies for access control. However, in practice, everything is not so simple. Today, some companies are ready to offer consumers access control using biometric technologies.

Traditional methods of personal identification, which are based on various identification cards, keys or unique data, such as, for example, a password, are not reliable to the extent that is required today.

A natural step in increasing the reliability of identifiers has been the use of biometric technologies for security systems.

The range of problems that can be solved using new technologies is extremely wide:

• preventing intruders from entering protected areas and premises by forgery, theft of documents, cards, passwords;
• restrict access to information and ensure personal responsibility for its safety;
• ensure access to critical facilities only for certified specialists;
• avoid overhead costs associated with the operation of access control systems (cards, keys);
• eliminate inconveniences associated with the loss, damage or simple forgetting of keys, cards, passwords;
• organize accounting of access and attendance of employees.

The development of technologies for recognizing images based on various biometric characteristics has been around for quite some time, beginning in the 60s. Our compatriots have achieved significant success in developing the theoretical foundations of these technologies. However, practical results were obtained mainly in the West and only “yesterday”. The power of modern computers and improved algorithms have made it possible to create products that, by their characteristics and ratios, have become accessible and interesting to a wide range of users.

The idea of ​​using individual characteristics of a person to identify him/her is not new.

Today, a number of technologies are known that can be used in security systems to identify a person by:

• fingerprints (both individual and the hand as a whole);
• facial features (based on optical and infrared images);
• iris;
• voice;
• other characteristics.

All biometric technologies have common approaches to solving the identification problem, although all methods differ in ease of use and accuracy of results.

Any biometric technology is applied in stages:

• scanning an object;
• extracting individual information;
• forming a template;
• comparing the current template with the database.

A biometric recognition system matches specific physiological or behavioral characteristics of a user to a given template. Typically, a biometric system consists of two modules: a registration module and an identification module.

The registration module trains the system to identify a specific person.

During the registration stage, a video camera or other sensors scan a person to create a digital representation of their appearance. Facial scanning takes about 20 to 30 seconds, resulting in several images.

Ideally, these images will have slightly different angles and facial expressions, which will allow for more accurate data. A special software module processes this representation and identifies characteristic features of the person, then creates a template. There are some parts of the face that hardly change over time, such as the upper contours of the eye sockets, the areas surrounding the cheekbones, and the edges of the mouth.

Most algorithms developed for biometric technologies allow for possible changes in a person's hairstyle, since they do not use areas of the face above the hairline for analysis. Each user's image template is stored in the biometric system database.

The identification module receives a person's image from a video camera and converts it into the same digital format in which the template is stored.

The received data is compared with a template stored in the database to determine whether the images correspond to each other. The degree of similarity required for verification is a certain threshold that can be adjusted for different types of personnel, PC power, time of day and a number of other factors.

Identification can be performed in the form of verification, authentication or recognition. Verification confirms the identity of the received data and the template stored in the database.

Authentication confirms the correspondence of the image received from the video camera to one of the templates stored in the database.

During recognition, if the received characteristics and one of the stored templates are the same, the system identifies the person with the corresponding template.

When using biometric systems, especially facial recognition systems, even when entering the correct biometric characteristics, the authentication decision is not always correct. This is due to a number of features and, first of all, to the fact that many biometric characteristics can change.

There is a certain degree of probability of system error. Moreover, when using different technologies, the error can vary significantly. For access control systems using biometric technologies, it is necessary to determine what is more important: not to let in “strangers” or to let in all of our own.”

An important factor for users of biometric technologies in security systems is ease of use. The person whose characteristics are scanned should not experience any inconvenience. In this regard, the most interesting method is, of course, facial recognition technology. However, in this case, other problems arise, primarily related to the accuracy of the system.

Despite the obvious benefits, there are a number of negative prejudices against biometrics, which often raise questions about whether biometric data will be used to spy on people and violate their privacy. Due to sensational claims and unfounded hype, the perception of biometric technologies differs sharply from the reality.

However, the use of biometric identification methods has become especially relevant in recent years. This problem became especially acute after the events of September 11 in the USA. The world community realized the degree of increase in the threat of terrorism throughout the world and the complexity of organizing reliable protection by traditional methods. It was these tragic events that served as a starting point for increased attention to modern integrated security systems. It is a well-known opinion that if control at airports were stricter, then accidents could have been avoided. Even today, finding those responsible for a number of other incidents could be significantly simplified by using modern video surveillance systems integrated with facial recognition systems.

Face recognition methods

Currently, there are four main face recognition methods:

• «eigenfaces»;
• analysis of «distinctive features»;
• analysis based on «neural networks»;
• method of «automatic face image processing».

All these methods differ in the complexity of implementation and the purpose of application.

«Eigenface» can be translated as «own face». This technology uses two-dimensional grayscale images that represent the distinctive characteristics of a face image. The «eigenface» method is often used as a basis for other face recognition methods.

By combining the characteristics of 100 120 «eigenface», a large number of faces can be reconstructed.

At the time of registration, the «eigenface» of each individual is represented as a series of coefficients. For the authentication mode, in which an image is used to verify identity, the «live» template is compared with the already registered template to determine the difference coefficient. The degree of difference between the templates determines the fact of identification.

The «eigenface» technology is optimal for use in well-lit rooms where it is possible to scan the face from the front.

The «feature analysis» technique is the most widely used identification technology. This technology is similar to the «Eigenface» technique, but is more adapted to changes in a person's appearance or facial expressions (smiling or frowning). The «feature» technology uses dozens of characteristic features of different areas of the face, taking into account their relative location.

The individual combination of these parameters determines the characteristics of each specific face. A person's face is unique, but quite dynamic, since a person can smile, grow a beard and moustache, wear glasses — all this increases the complexity of the identification procedure. Thus, for example, when smiling, there is some displacement of the parts of the face located near the mouth, which in turn will cause a similar movement of adjacent parts.

Taking into account such shifts, it is possible to uniquely identify a person even with various facial expression changes.

Since this analysis examines local areas of the face, permissible deviations can be within the range of up to 25° in the horizontal plane, and approximately up to 15° in the vertical plane, and requires fairly powerful and expensive equipment, which accordingly reduces the extent of distribution of this method.

In a neural network-based method, the characteristic features of both faces — the registered and the one being checked — are compared for coincidence.

Neural networks use an algorithm that establishes a correspondence between the unique parameters of the face of the person being checked and the parameters of the template in the database, using the maximum possible number of parameters.

As the comparison proceeds, discrepancies between the face being checked and the template from the database are determined, then a mechanism is launched that, using appropriate weighting factors, determines the degree of compliance of the face being checked with the template from the database. This method increases the quality of face identification in complex conditions.

The «automatic face image processing» method is the simplest technology, using distances and the ratio of distances between easily identifiable points on the face, such as the eyes, the end of the nose, and the corners of the mouth. Although this method is not as powerful as «eigenfaces» or «neural network», it can be used quite effectively in low-light conditions.

Face recognition systems available on the market

Today, a number of commercial products have been developed for face recognition. The algorithms used in these products are different, and it is still difficult to assess which technology has the advantage.

The current leaders are the following systems: Visionic, Viisage, and Miros.

• Visionic's FaceIt app is based on a local feature analysis algorithm developed at Rockefeller University. A commercial company in the UK has integrated FaceIt into a television anti-crime system called Mandrake. The system looks for criminals based on video data from 144 cameras connected in a closed network. When an identity is established, the system notifies a security officer. Visionic's representative in Russia is DanCom.
• Another leader in this field, Viisage, uses an algorithm developed at MIT. Commercial companies and government agencies in many US states and other countries use Viisage's system along with identification cards such as driver's licenses.
• ZN Vision Technologies AG (Germany) offers a number of products on the market that use facial recognition technology. These systems are presented on the Russian market by Soling.”
• Miros’ TrueFace facial recognition system uses neural network technology, and the system itself is used in the Mr.Payroll cash dispensing complex and is installed in casinos and other entertainment venues in many US states.

In the USA, independent experts conducted comparative testing of various facial recognition technologies. The results of the testing are presented below.

Fig. 1. Comparative analysis of facial recognition efficiency in different systems

In practice, when using facial recognition systems as part of standard electronic security systems, it is assumed that the person to be identified is looking directly into the camera. Thus, the system works with a relatively simple two-dimensional image, which significantly simplifies the algorithms and reduces the intensity of calculations. But even in this case, the recognition task is still not trivial, since the algorithms must take into account the possibility of changing lighting levels, changes in facial expression, the presence or absence of makeup or glasses.

The reliability of the facial recognition system depends greatly on several factors:

• Image quality. The probability of error-free operation of the system is significantly reduced if the person we are trying to identify is not looking directly at the camera or is photographed in poor lighting.
• Relevance of the photograph entered into the database.
• Size of the database.

Facial recognition technologies work well with standard video cameras that transmit data and are controlled by a personal computer, and require a resolution of 320-240 pixels per inch with a video stream rate of at least 3-5 frames per second.

For comparison, acceptable quality for a video conference requires a video stream rate of 15 frames per second.

A higher video stream rate with a higher resolution leads to improved identification quality.

When recognizing faces from a long distance, there is a strong relationship between the quality of the video camera and the identification result.

The volume of databases using standard personal computers does not exceed 10,000 images.

Conclusion

The methods of facial recognition proposed today are interesting and close to widespread implementation, but it is not yet possible, as in the movies, to trust the opening of the door only to facial recognition technology. It is good as an assistant for a security guard or other access control system.

This method is used in many situations when it is necessary to verify that the presented document actually belongs to the person who presented it. This happens, for example, at an international airport, when a border guard compares the photo on the passport with the face of the passport owner and decides whether it is his passport or not.

The computer access system operates according to a similar algorithm. The only difference is that the photograph is compared with a template already stored in the database.

Technologies based on facial recognition in infrared light have already appeared.

The new technology is based on the fact that a thermal image created by the radiation of heat from the blood vessels of the face, or, in other words, a thermogram of a person's face, is unique for each person and, therefore, can be used as a biometric characteristic for access control systems.

This thermogram is a more stable identifier than facial geometry, since it is almost independent of changes in a person's appearance.