Comparison of several biometric technologies

Biometric technology recognizes the identity of individuals based on their own characteristics such as fingerprints, sounds, etc. Currently, there are many biometric technologies available for identity authentication. Here, we describe how most popular biometric technologies work, and briefly comment on their functions of capturing images, extracting features, comparing and comparing.

Iris recognition technology

The iris is a textile-like ring of various colors in the pupil of the eye Structure, it is claimed that no two irises are the same. The iris scanning security system includes a fully automatic camera to find your eyes and start to focus when the iris is found. It is impossible to deceive the system by blinking your eyes.

Advantages of iris technology

· Convenient for users;
· Probably the most reliable biometric technology, although it has not been tested;
· Only the user is in front of the device without physical contact.

Disadvantages of iris technology

One of the most important shortcomings is that it has not conducted any tests. The current iris recognition system only uses statistical principles to conduct small-scale tests, and has not conducted real-world unique certification tests;
· It is difficult to downsize the image acquisition device;
· Expensive cameras are required due to the need for focusing. The minimum offer for such a camera is $ 4000;
· The lens may distort the image and greatly reduce the reliability;
· Dark eyes are extremely difficult to read;
· Need a better light source.

  Retina recognition technology

· The retina is also a feature used for biometric recognition. Some people think that the retina is the only biometric feature than the iris. Retinal recognition technology requires laser light to illuminate the back of the eyeball to obtain the uniqueness of the retina.

Advantages of retina technology

· The retina is an extremely fixed biological feature, because it is "hidden", so it is not likely to wear, aging or disease;
· The user does not need to make direct contact with the device;
· It is the most difficult system to deceive because the retina is invisible, so it cannot be forged.

Disadvantages of retina recognition

· Retina technology has not been tested.
· Obviously, retinal technology may cause health damage to users, which requires further research;
· Retina technology is not attractive to consumers;
· It is difficult to further reduce its cost.

face recognition

Facial recognition technology recognizes facial features and their relationships. Recognition technology is very complicated based on these unique features. This requires artificial intelligence and machine knowledge learning systems. Two technologies used to capture facial images Standard video and thermal imaging technology. Standard video technology takes a face image or a series of images through a standard camera. After the face is captured, some core points are recorded, for example, the positions of the eyes, nose, and mouth and the relative positions between them are recorded and then formed Template; thermal imaging technology generates facial images by analyzing the heat generated by the blood of the capillaries of the face. Unlike video cameras, thermal imaging technology does not need to be in a better light source condition, so it can be used even in the dark . An algorithm and a neural network system plus a conversion mechanism can turn a fingerprint image into a digital signal, which eventually produces a matched or unmatched signal.

Advantages of facial recognition

· Facial recognition is non-contact, users do not need to directly contact the device;
· Although the desktop video camera can be used, only a more advanced camera can effectively capture facial images at high speed;

Disadvantages of facial recognition

· The position of the user's face and the surrounding light environment may affect the accuracy of the system;
· Most people who study biometrics recognize that facial recognition is the least accurate and the most easily deceived;
· The improvement of facial recognition technology depends on the improvement of feature extraction and comparison technology, and the equipment for collecting images will be much more expensive than its technology;
· For changes in the human face such as hair, accessories, aging, and other changes that may need to be compensated by artificial intelligence, the machine learning function must constantly compare the previously obtained images with the current ones; And make up for small differences;
· It is difficult to further reduce its cost, we must sell high-quality equipment at an expensive cost.

Signature recognition

· Signature has been used as a means of identity authentication for hundreds of years, and we are all familiar with signing in the bank's format form as a sign of our identity. Digitizing a signature is a process that measures the image itself and the action of the entire signature—the different speeds, sequences, and pressures between each letter and between letters. Signature recognition is the same as voice recognition, and it is a behavior determination.

   Advantages of signature recognition

· The use of signature recognition is more easily accepted by the public and is a recognized identification technology.

   Disadvantages of signature recognition

· As experience grows, temperament changes and lifestyle changes, the signature will also change with it;
· In order to deal with the inevitable natural change of signatures, we must compromise on security;
· Because the signature is not fast, we cannot use it on the Internet;
· The structure of the handwriting board used for signature is complicated and expensive, because there is a big difference from the resolution of the touchpad of the notebook computer, it is technically difficult to combine the two;
· It is difficult to miniaturize its size.

   Voice recognition technology

· Like signature recognition, voice recognition is also a behavior recognition technology. Voice recognition equipment continuously measures and records the waveform and changes of voice. Voice recognition is based on accurately matching the voice collected on the spot with the registered voice template.

Advantages of voice recognition

· Voice recognition is also a non-contact recognition technology that users can accept naturally.

Disadvantages of voice recognition

· As with other behavior recognition technologies, the range of sound changes is too large, so it is difficult to perform some precise matching;
· The sound will affect the result of collection and comparison with the change of volume, speed and sound quality (for example, when you have a cold);
· With the development of technology, you may be able to perceive and refuse the sound of the recording. However, at present, it is easy to deceive the voice recognition system with the sound recorded on the tape.
· High-fidelity microphones are very expensive.

  Fingerprint recognition system

As a recognition technology, fingerprint recognition has a long history and has a solid market backing. According to ordinary people's views, fingerprint recognition technology analyzes the global characteristics of fingerprints and local characteristics of fingerprints, such as ridges, valleys and end points, and forks. Point or divergence point, the feature values ​​extracted from the fingerprint can be very detailed in order to reliably confirm the identity of a person through the fingerprint.

On average, each fingerprint has several unique and measurable feature points, and each feature point has about seven features. Our ten fingers produce at least 4900 independently measurable features. Reliable identification method. Here, we generally give the process of acquiring images, extracting features and comparing in fingerprint recognition.

Optical technology and capacitance technology in image acquisition

The two technologies used to collect fingerprint images are mainly optical technology and capacitive technology. Optical technology requires a light source to reflect from the prism and press a finger on an imaging head, and the light illuminates the fingerprint to collect the fingerprint.

Using semiconductor technology of capacitive technology, the ridges and valleys of the finger pressed on the collection head generate different capacitances between the finger skin and the chip. The chip obtains a complete fingerprint by measuring different capacitance fields in the space.

Disadvantages of capacitive technology

Because the chip of capacitive technology is expensive, the size of the chip and the finger are already expensive, so several companies try to introduce a chip that can provide a smaller size than the fingerprint to collect only part of the fingerprint for verification. To use this collection method, the user must be accurate Put your finger to ensure correct reading. And this will inevitably make the reading head difficult to use. Another disadvantage of using this small chip is that only a part of the fingerprint must not be collected and the entire fingerprint is much more reliable for comparison.

Another disadvantage of the capacitive collection head is that it is susceptible to interference, from the interference of the 60HZ cable to the interference of user contact, the electrical interference inside the fingerprint collector, etc.

The last problem of the capacitive collection head is reliability. Whether it is static interference, salt in sweat or other stolen goods and finger wear will make the collection head difficult to read fingerprints.

In fact, so far, the optical collection head provides a more reliable solution. By improving the original optical imaging technology, the new generation of optical fingerprint collector even dwarfs the capacitive solution with impeccable performance and relatively low price.

   Advantages of fingerprint recognition

· Fingerprints are unique features of the human body, and their complexity is sufficient to provide sufficient features for identification;
· If we want to increase reliability, we only need to enroll more fingerprints, identify more fingers, up to ten, and each fingerprint is unique;
· The fingerprint scanning speed is very fast, very convenient to use;
· When reading fingerprints, users must touch their fingers with the fingerprint collection head. Direct contact with the fingerprint collection head is the most reliable method for reading human biological characteristics. This is also a major reason why fingerprint recognition technology can occupy most of the market.
· The fingerprint collecting head can be more miniaturized and the price will be lower.

  Disadvantages of fingerprint recognition

· The fingerprints of some people or groups are very difficult to image because they have few fingerprint characteristics;
· In the past, because of the use of fingerprints in criminal records, some people were afraid of "recording fingerprints". However, in fact, the current fingerprint authentication technology can guarantee not to store any data containing fingerprint images, but only store encrypted fingerprint characteristic data obtained from the fingerprint.
· Every time the fingerprint is used, the fingerprint of the user will be left on the fingerprint collection head, and these fingerprints may be used to copy the fingerprint.
· It can be seen that the fingerprint identification technology is currently the most convenient, reliable, non-invasive and inexpensive biometric technology solution, and has great potential for application in the broad market.

The importance of software and firmware in fingerprint recognition technology

Although the price, size and hardware design are very important for the fingerprint identification system, it only determines whether the entire system is robust, but the importance of firmware and software in the entire system is not lower than that of hardware, especially for those who want to occupy the market Product.

The firmware in the fingerprint collection head is responsible for processing the image and connecting with the PC. In most systems, the firmware is relatively simple. It constantly transmits data to the computer. However, there are many problems in doing so. First The main problem is that the transmitted data can be easily recorded and reused, which makes the system potentially dangerous. Another problem is that we must supply power to the fingerprint collection head. The computer must continuously capture images to determine whether there is a fingerprint pressed up, so as to capture the fingerprint image at the most appropriate time. Through the designed firmware to deal with these problems, the performance of the entire system can be improved; using the USB interface is a good interface solution at present, which can provide power, bandwidth and plug-and-play functions; fingerprints transmitted to the computer need to be performed Encryption to ensure security; and, in the state of not reading the fingerprint, the firmware should go to a low-power state.

After the host computer securely obtains the fingerprint image from the computer, the recognition algorithm proceeds to the next verification process. Fingerprints are such reliable biometrics that they can be compared with very little information. However, this feature of fingerprints is not reflected in most fingerprint recognition algorithms. Most systems require users to press all their fingers, and users must press their fingers very carefully. If the position of the fingerprint is not correct or the fingerprint quality is not high, the verification cannot be performed, and the user must press the finger again, and such a product cannot stand on the market.

Therefore, a good system should be easier to use and reliable, users do not have to worry about the placement of fingerprints, the algorithm must support 360-degree rotation and torn fingerprints. The user only needs to gently press his finger without worrying about whether it is in a proper position or pressing only a part. For the pressure, rotation, quality of the fingers, and the dust and mist of the collection head, the system should be able to solve it well.

For surveillance and non-surveillance operations, one-to-one comparison and one-to-many comparison using fingerprint recognition technology must first understand the difference between surveillance and non-surveillance comparison. In the traditional criminal fingerprint database, fingerprints are under surveillance The acquisition and comparison is based on multiple comparisons in the same fingerprint database. In order to introduce fingerprint recognition algorithms to the market, fingerprint reading must be based on unsupervised status. In this way, you only need to press lightly without waiting for the fingerprint image to reach the best. In addition, when processing e-commerce, the data will be compared with the registered fingerprint. In most cases, it is a one-to-one comparison instead of One-to-many comparison. Unsupervised acquisition and one-to-one comparison algorithms and one-to-many comparison algorithms have good countermeasures in the system. The unsupervised mode must enable the comparison algorithm to handle poor-quality fingerprints and the algorithm must compare with the monitoring Be reliable in the state.