Biometric identification systems: reality, not utopia.

Biometric identification systems: reality, not utopia.

Biometric identification systems: reality, not utopia

The idea of ​​using human biometric parameters for identification purposes rightfully belongs to science fiction writers (Robert Heinlein, Isaac Asimov, etc.). Scientists, in turn, took on the task of making the dream a reality by creating a special science called biometrics. Methods of measuring various parameters of a person in order to establish similarities or differences between people and to distinguish one specific person from many other people — all this includes biometrics. The applied aspect of biometrics is biometrics, which uses unique human characteristics — biometric characteristics — when creating various automatic access control systems.: papillary pattern of a finger; shape of a hand; pattern of the iris of the eye; voice parameters; facial features; thermogram of a face (blood vessel diagram); form and method of signature; fragments of genetic code. Biometric identification systems work on the following principle: first, a person’s biometric characteristics are entered into a sample; for greater accuracy, several samples are often made. The obtained data is processed and encoded digitally. Biometric security systemsoperate in two modes: verification and identification. Verification is the process of comparing the received data with a biometric template. Verification can be done using a smart card, username, or identification number. Identification is a one-to-many comparison process: after “capturing” biometric data, a connection is made to a biometric database to determine the identity. Biometric identification is successful if the biometric sample is already in the database. First biometric identificationwas based on registration. During registration, biometric information from the individual was stored. Subsequently, the biometric information was recorded and compared with information obtained earlier. Biometric Identification Systemsare based on several stages: sensor, block and the third block. The sensor is an intermediate stage that links the real parameters with the system, the sensor must receive all the necessary data. In most cases, these are images, but the sensor can also work with other data in accordance with the desired characteristics. The block is the second stage — it carries out all the necessary preliminary processes: it must remove all the «excess» from the sensor to increase the sensitivity at the input. In the third block, the necessary data is extracted. A vector of values ​​or an image with special properties is used to create a template. A template is a synthesis of the relevant characteristics extracted from the source. Elements of biometric measurement that are not used in the comparison algorithm are not saved in the template in order to reduce the file size and protect the identity of the registrant, making it impossible to recreate the original data from the information in the template. The registration template is stored in the access card and/or in the biometric system database.

Biometric technologies

Characterized by the speed and simplicity of identification. The main types of identification are listed below: Identification by fingerprints is the most common and developed biometric technology. Up to 60% of biometric devices use it. Three main methods are used to take a fingerprint and then recognize the sample: optical, semiconductor and ultrasonic.

• Optical scanner

  works using an optical method of obtaining an image. FTIR scanners use the effect of attenuated total internal reflection. In this case, the finger is illuminated, and a special camera is used to receive the light image. A fiber optic optical scanner is a fiber optic matrix, each fiber of which is equipped with a photocell. The principle of obtaining a pattern is the recording of residual light passing through the finger to the surface of the scanner. Electro-optical scanners — a special electro-optical polymer with the help of a light-emitting layer illuminates the fingerprint, which is recorded using a special camera. Contactless optical scanner: the finger is placed on a special hole in the scanner, several light sources illuminate it from below. The reflected light is projected onto the camera through a collecting lens. There is no contact with the surface of the reading device. Roller scanners — when scanning, the user rolls a small transparent cylinder with their finger. A static light source, a lens and a camera are located inside it. During the finger movement, a series of pictures are taken of the papillary pattern in contact with the surface.

• Ultrasonic scanners

  The surface of the finger is scanned using ultrasonic waves. The distance between the wave source and the depressions and protrusions of the finger surface is measured by the reflected echo.

• Face identification

  These include a method of capturing an image and recognizing a face automatically in real time. It is similar to the method of recognizing people by face.

• Iris identificationA person can be identified by the iris, even if he or she is wearing glasses or contact lenses. Iris identification has been tested on various ethnic groups and nationalities and has proven its reliability and accuracy; it has been working stably for several years now.
• Voice identificationUses the unique acoustic features of speech. They reflect human anatomy: the size and shape of the larynx and mouth, as well as acquired properties such as voice volume, speed and manner of speaking. Voice identification involves asking a person to answer two or three questions that are easy to remember. For example: last name, first name, patronymic; date of birth. Some modern systems create a voice model and can match it with any phrase spoken by a person.

Signature identification and other behavioral biometric technologies are based on measuring a person's behavioral characteristics. These include writing speed, pressure and tilt of letters, pen movement when signing or writing text. When identifying by the manner of working with a keyboard, such indicators as the dynamics of pressing keys, rhythm, and the user's manner of pressing keys are taken. Unfortunately, despite the wide range of possible applications, these methods have not yet received sufficiently widespread distribution.