1. Fingerprint identification: There are many ways to realize fingerprint identification. Some imitate the traditional methods of comparing local details of fingerprints by public security departments; Some are directly identified by all features; Others use more unique methods, such as the ripple edge pattern of fingerprints and ultrasound. Some devices can measure finger fingerprints in real time, while others cannot.
Among all biometric technologies, fingerprint identification is the most widely used one at present. Fingerprint identification is more suitable for indoor security system, because it can provide sufficient conditions for users to explain and train, and the operating environment of the system is controllable. Because of its relatively low price, small size and easy integration, almost all applications in workstation security access control system are fingerprint identification.
2. Palm geometry recognition: Palm geometry recognition is recognized by measuring the physical characteristics of the user's palm and fingers, and advanced products can also recognize three-dimensional images. As a mature method, geometric recognition of palm is not only good in performance, but also convenient to use. Suitable for a large number of users, or users do not often use it, but it is easy to accept when using it.
3. Speech recognition: Speech recognition is a technology to recognize by analyzing the physical characteristics of the user's voice. Although some speech recognition products have entered the market now, they are not convenient to use, mainly because of the great variability of sensors and voices. In addition, compared with other biometric technologies, it uses more complicated steps and is more inconvenient in some occasions. A lot of research work is under way, and we believe that speech recognition technology will make great progress.
4. Retinal recognition: Retinal recognition uses a low-intensity light source emitted by optical equipment to scan unique patterns on the retina. There is evidence that retinal scanning is very accurate, but it requires users to look at the receiver and stare at a point. This is inconvenient for people who wear glasses, close to the receiver and uncomfortable.
So although the retina recognition technology itself is very good, the user's acceptance is very low. Therefore, this kind of product is still a non-mainstream biometric product, although it was redesigned in the 1990s to enhance connectivity and improve the user interface.
5. Iris recognition: Iris recognition is a technology that causes little interference to people in biometric identification related to eyes. It uses quite common camera elements and does not require users to touch the machine. In addition, it can achieve higher template matching performance. Therefore, it attracts the attention of all kinds of people. In the past, iris scanning equipment had no advantages in simplicity of operation and system integration, and we hope that the new products can be improved in these aspects.
6. Signature identification: Signature identification has advantages over other biometric technologies in application. People are used to signing as a way to confirm their identity in transactions, and its further development will not make people feel much different. Practice has proved that signature recognition is quite accurate, so the signature can easily become an acceptable identifier. However, compared with other biometric products, there are few such products now.
7. Facial recognition: This is a very attractive technology, and its performance is often misunderstood. There are often some exaggerated remarks about facial recognition, but it is difficult to achieve in practice. It's one thing to compare two still images, but it's quite another to find and confirm someone's identity in a crowd without attracting others' attention.
Some systems claim to be able to do the latter, but they are actually doing the former, but they are not biometrics. It is easy to understand the attraction of facial recognition from the user's point of view, but people's expectations of this technology should be realistic. Face recognition is rarely successful in practical application. But once the technical obstacles are overcome, it will become an important biometric method.