As technology continues to evolve, various types of new biometric technologies are emerging and applied to all aspects of our lives. Many people through the phone to unlock, payment, and even bank client login has a lot of people set a fingerprint login, or even face recognition login. More and more new biometrics are brought out. How many technologies are before us and which one is our favorite dish?
Signature recognition, this is an old antiques
You go to the mall shopping, business card after the card will let you sign a list, which uses the signature recognition. Signature recognition is based on the signature of the font for authentication, which is an old antiques, when the name can specify their own identity has emerged, of course, then it is not biometric.
Everyone has their own unique writing style, so signature recognition can be used as a kind of biometrics, which belongs to the behavior test. Digitizing a signature is a process of measuring the image itself and the entire signature-the different speeds, sequences, and pressures between each letter and the letter. Signature recognition is divided into offline signature recognition and online signature recognition. Online recognition is through the handwriting board to collect the signature of the writer's signature, and some systems can also capture the pressure, hold pen perspective and other data, the recognition rate is very high, but because of human writing and writing dynamic stereotypes are not fixed, a certain range , So just from the glyph, it may not be able to distinguish between the real signature and forged signatures. Signed offline is even more unsafe.
Sound recognition, feeling a take you may not be yours
Sometimes sound recognition is not as secure as signing. Recognition of sound also belongs to behavior recognition technology, recognition devices constantly measure and record the waveform and change of sound. Sound recognition is based on precisely matching sound collected at the scene with registered sound templates.
The advantage is that voice recognition is a non-contact recognition technology that users can naturally accept, nothing more. Shortcomings are more, such as the scope of the sound changes is too large, it is difficult to accurately match; such as flu, nose is not ventilated, this time you can hardly make the recognition system to know you are you. Even taking a tape may break it. Do not ask me why I do not have a high-fidelity microphone, because it is expensive, recording it I did not want to buy a little more than a small identity authentication. This biometric, entertainment and entertainment on the line, do not take it seriously.
Iris recognition, a contact lens with iris information is really possible to get the recognition system
We often see some tech-based or high-IQ criminal films that use iris recognition as an important means of identifying people, who are then hacked by some high-IQ criminals wearing contact lenses that have stolen iris information. Unfortunately, in theory, it can be done.
The iris is a muscle that regulates the pupil size outside the eye and controls the amount of light that enters the eye, which is a colored part of the eye based on the amount of melatonin. Iris recognition technology is through the identification of human iris to confirm the identity is a biometric authentication technology.
Before the iris recognition, it needs to be located by the landmark features and use these features and unique shape to iris imaging, feature separation and extraction. Based on iris imaging, two-dimensional Gabor waves are screened and plotted as phasor. Phasor information includes direction and spatial frequency (image content) and image position. These phasor information are used to draw "iris code" Certification.
The iris recognition technology is almost the most accurate biometrics, as described below: The probability that 75% of two different iris information matches the information is 1: 106, etc. Error rate: 1: 1200000 Two different irises produce the same iris The possibility of code is 1: 1052.
High precision, without physical contact is its advantage, which is why banks and other static terminal fancy it. However, for the Internet of Things era, the use of iris recognition on mobile terminals has some limitations. For example, iris recognition products are expensive and complex to implement. Very demanding on the iris image collected. Iris's refusal rate is high, when the refusal of the situation, it is difficult to artificially determine the employee's iris is true or false, unless you find expert identification. Difficult to identify people with eye problems is also a disadvantage, and it is really possible to be faked.
Fingerprinting, a biometrics transmitted from the Tang Dynasty in China to the world
In 1927, Robert Heindel in Germany concluded in "Fingerprint Identification" that Jia Gongyan in the Tang Dynasty in China was the first scholar in the world to propose using fingerprinting. Ancient China in this regard is not without (yin).
Fingerprint recognition as a recognition technology has a long history, has a solid market backing, according to the general view, fingerprint recognition technology by analyzing the global characteristics of fingerprints and fingerprints of local features, such as ridge points, valleys and end points, bifurcation points The eigenvalues ​​extracted from fingerprints can be very detailed and can effectively confirm a person's identity.
On average, each fingerprint has several unique measurable features, each featuring about seven features, and our ten fingers producing at least 4,900 individually measurable features.
Today's fingerprint image acquisition technology mainly for radio frequency fingerprint identification technology. RF sensor technology is through the sensor itself emits a small amount of RF signals, penetrate the epidermis of the fingers to control the inner lines to get the best fingerprint image. Therefore, the fingers, sweat fingers, dirty fingers and other difficult fingers through up to 99%, anti-counterfeit fingerprints ability, fingerprint sensor identification principle only to the human dermal skin reaction, fundamentally eliminate the artificial fingerprints.
The recent fire in mobile payment biometric technology, because of its high security, unique biological characteristics, the core chip continues to shrink and so on, has been widely used in flat, mobile phones and even watches and other mobile terminals, and will in the future mainstream biometric Market place. Put your finger on the screen of your cellphone, and you can confirm the payment with one stroke. Combining safety with convenience. However, it also has a formidable opponent, face recognition.
Face recognition, Ali Tencent recognized biometric methods
As one of the mainstream biometric identification methods in the future, the biggest advantage of face recognition is that it requires almost no hardware, and even a few hundred yuan of mobile phone camera can meet, leaving only a software problem. The result for the future is that when you checkout, you look at the camera at the cashier's desk and take a look at the right. The payment is completed. This authentication method in the future mobile payment era with high convenience, coupled with very low hardware requirements, enough to make Ali Tencent optimistic about its future.
The classical automatic face recognition process is divided into three steps: face detection, facial feature point location, feature extraction and classifier design. Before the emergence of deep learning, face recognition methods are generally divided into high-dimensional artificial features extraction and dimension reduction two steps. 2014 years
Cnc Milling Machine,Cnc Mill,Desktop Cnc Mill,Mini Cnc Mill
Taizhou Jiangzhou CNC Machine Tool Manufacturing Co.,Ltd , https://www.tzjzmachine.com