At the time of payment, both the password and the recently popular fingerprint have a common purpose, which is to prove that you are you. Professionally speaking, it is called identification.
Therefore, what fingerprints do is no different from airport staff using ID cards to compare your looks. In fact, the accuracy of fingerprint recognition is lower than the accuracy of face-to-face verification of ID cards. To put it more straightforward, fingerprint recognition is a method of authentication that sacrifices certain accuracy and greatly exchanges for ease of use.
Although we are still satisfied with the accuracy and ease of use of fingerprint recognition, is fingerprint recognition really the best solution in the future? In the ant gold service identity certification forum, all the big cattle gave a lot of "black technology."
Fingerprint recognition limitThe basic principles of the two most common fingerprint recognition methods at present are ultrasound and capacitance. However, each of these two sensors has one fear:
Ultrasound is afraid of "dry", the recognition rate of dry fingers is low, especially in the dry winter in the north, the ultrasonic-based fingerprint unlocking is maddening.
The capacitor is afraid of "wet", and the recognition rate for wet hands is low, and there is no solution state for fingers that reach a certain humidity. Especially in the rainy summer, or after the user has just bathed.
Although most people reading this article can use fingerprint payment without any obstacles, according to the data provided by several industry giants, nearly 5% of people are born to be the "buster" of this identification method. E.g,
Some people's fingers are very light fingerprints, some people sweat sweating nervously, and some people are extremely dry. These "fantastic" fingers continue to challenge the physical limits of fingerprint recognition technology.
Others are engaged in agriculture or heavy physical labor, and fingerprints often carry dirt or oil. This situation will make all fingerprint identification techniques dumb.
In fact, users who are not suitable for fingerprint recognition are more likely than most people think. For payments, unlocking mobile phones, which almost everyone needs, fingerprints are clearly not the best solution.
[Region for human eye identification: Pupil, pupil; Iris, iris; Scler, sclera]
Iris loveWhen it comes to identity authentication, a technology that is hard to ignore is iris recognition.
Iris recognition is the “Gao Fu Shuai†in the identity authentication technology. The recognition accuracy is extremely high, and the error rate is only one in 100,000. But at the same time, it has the highest requirements for hardware, algorithms, and recognition conditions.
In simple terms, this part of the area outside the pupil and the white of the eye is the location of the iris recognition.
Zhongke Hongba is a company specializing in iris recognition technology. Its CEO Ma Li said:
Since the age of one person, the shape of the iris has stabilized, and the shape of the iris is different in different individuals, so it can be used to identify identity. At present, there are many professional authentication scenes that use iris recognition, but in the life of ordinary people, iris recognition is still somewhat strange.
The reason why iris recognition is not yet big is because this technology requires a specialized iris acquisition camera. At present, this kind of dedicated camera is obviously not popular in the popular mobile phone. The calculation of iris comparison requires complex algorithms and has certain requirements on the computing power of the device. In addition, to achieve high-availability iris recognition on mobile devices, the technology needs to be improved.
In horsepower, although there are many paths to implementation, "it still takes a little time."
[Comparison of security and user experience of mainstream identity authentication technology]
Face recognition for God horse?Many people have used face recognition technology in Alipay or other scenarios.
Li Liang, an expert from Ant Financial, said:
In fact, face recognition technology has surpassed people's recognition of faces. Talented people can remember 500 familiar faces, while ordinary people can remember 300-400 familiar faces. Within this range, people may perform better than algorithms. However, people will have extremely obvious "face blindness" for unfamiliar faces. So in large-scale face recognition, people are definitely not the opponent of the machine.
Let's look at a set of interesting comparison data:
The misrecognition rate of fingerprint recognition is one in 20,000, the misidentification rate of iris recognition is one in 100,000, and the misrecognition rate of face recognition is one in ten thousand.
That is to say, if only the face recognition locking device is used, one out of every 10,000 people can fool face recognition and enter your mobile phone.
[Deceive face recognition in various wonderful ways]
The rate of misrecognition is twice that of fingerprint recognition. The false recognition rate is not high, but it is obvious that face recognition is not as reliable as fingerprint recognition. However, in real-life applications, face recognition has many advantages.
Li Liang said:
In real life, many of the IDs used by people, such as ID cards, driver's licenses, and social security cards, are the subject of forensic identification. The relevant state departments have a very complete citizen face information database, which can provide external inquiry services. This infrastructure can bring a huge benefit, that is, all people who use face recognition do not need to enter face information in advance. This is a major consideration in the use of face recognition technology by Ant Financial.
Without registration, direct identification, this really makes people who are getting lazy feel happy. However, face recognition in the application scenario, but it needs to face many challenges:
1. Illumination: When using face recognition on the mobile side, people are often in a strange lighting environment, such as a dimly lit bedroom or toilet, a light and shadowy bar. “I can't see my face†is a huge challenge for face recognition.
2, posture: In many scenes, people do not absolutely match the face recognition, 45 degrees look up at the sky often makes the face recognition system want to cry without tears.
3, expression: When people take photos, they are generally a pair of iron-green faces, but when they are shopping, they are all smiling.
4, makeup: If you have seen the online circulation of girls before and after makeup, you should know what I am talking about.
5. Age: As the age increases, the person's appearance will also change.
Li Liang said that for the first three challenges, Ta can be matched by prompting the user, but for the latter two challenges, it can only be improved by big data and algorithm optimization. (When a beautiful woman performs face recognition, the system prompts: Please remove the makeup and try again, what will happen...)
Wall adapter power supply-5v dc 2a, -9vdc 650ma, -12vdc 600ma. Wall adapter can be used for cctv, led light, mini electronic device.We can meet your specific requirement of the products, like label design. The plug type is US/UK/AU/EU. The material of this product is PC+ABS. All condition of our product is 100% brand new.
Our products built with input/output overvoltage protection, input/output overcurrent protection, over temperature protection, over power protection and short circuit protection. You can send more details of this product, so that we can offer best service to you!
Wall Charger,Wall Power Supply,Wall Wart Power Charger,Wall Mount Charger
Shenzhen Waweis Technology Co., Ltd. , https://www.waweis.com