The unit that used to punch cards is now a fingerprint machine. I don't know what this means.

Hello! You can cheat by punching in, or you can ask others to do it for you. Fingerprints can't be fooled, because everyone's fingerprints are different. Fingerprint machine is very sensitive to recognition. You can't recognize hand peeling in spring, and it's annoying! I copied the following information, which is very authoritative. Can be used as a reference!

fingerprint

A fingerprint on a finger represents a person's identity. Mayer first proposed in 1788 that no two people's fingerprints are exactly the same, Purkinie first classified fingerprint patterns into nine categories in 1823, and Henry put forward the theory of fingerprint minutiae recognition in 1889, which laid the foundation for modern dactylography. But the manual comparison method is inefficient and slow. In 1960s, computer image processing and pattern recognition methods were used to analyze fingerprints, which is the Automatic Fingerprint Identification System (AFIS)[ 1]. In the late 1970s and early 1980s, automatic fingerprint identification system (police? AFIS pass. AFIS) put into practical use. In the 1990s, AFIS entered the civil service, which is the so-called civil automatic fingerprint identification system (CIVIL? AFIS pt? AFIS). This paper tries to start with the analysis of fingerprint characteristics, and expounds the principle and method of fingerprint as human identification, the main technical indicators and testing methods of fingerprint identification, as well as the reality and reliability of practical application [2-4].

1 Principles and methods of fingerprint identification

1. 1 Characteristics and classification of fingerprints

Fingerprint identification is an ancient subject, which is developed based on the relatively stable and unique statistical results of human fingerprint characteristics. In practical application, according to different needs, human fingerprint features can be divided into permanent features, non-permanent features and vital features [5].

Permanent features include detailed features (center point, triangle point, end point, bifurcation point, bridge point, etc. ) and auxiliary features (particle type, particle density, particle curvature and other elements), which will never change in a person's life, and are most obvious and evenly distributed in the typical area in front of the finger [1]. Detail features are the basis of accurate fingerprint comparison, while pattern features and texture features are the important basis for fingerprint classification and retrieval. According to their different shapes, the tattoo features of human fingerprints can usually be divided into three types: bow type, dustpan type and bucket type, and nine types: solitary type, tent type, positive dustpan type, inverted dustpan type, ring type, spiral type, bag type, double dustpan type and miscellaneous type [1]. Texture features are composed of texture parameters such as average particle density, particle density distribution, average particle curvature and particle curvature distribution. Texture features are mostly used for multi-dimensional classification and retrieval of computer fingerprint identification algorithms.

Non-permanent features are fingerprint features composed of isolated points, short lines, folds, scars and the resulting breakpoints, forks and other elements, which may be generated, healed, developed or even disappeared [1].

The life characteristics of fingerprints are closely related to the existence of the life of the tested object. However, the relationship and law between it and human life phenomena still need further understanding. At present, it has become one of the hot spots in the application of modern civil fingerprint identification.

1.2 Principle and method of fingerprint identification

Fingerprint identification technology mainly involves four functions: reading fingerprint images, extracting features, saving data and comparing. The image of human fingerprint is read by fingerprint reading equipment, and then the original image is preliminarily processed to make it clearer, and then the fingerprint feature data is established by fingerprint identification software. The software finds data points called "minutiae" from fingerprints, that is, the coordinate positions where fingerprint lines diverge, terminate or surround, and these points have more than 7 unique features at the same time. Usually there are 70 nodes on a finger, so this method will produce about 490 data. These data are often called templates. Through computer fuzzy comparison, compare the templates of two fingerprints, calculate their similarity, and finally get the matching result of two fingerprints [5-6]. Acquisition equipment (i.e. image acquisition equipment) is divided into several categories: optical, semiconductor sensors and others.

2. Main indexes and test methods of fingerprint identification technology.

2. 1 Accuracy of the algorithm

The performance index of fingerprint identification system depends largely on the performance of the adopted algorithm. In order to quantitatively express its performance, the following two indicators are introduced.

False Rejection Rate (FRR): refers to the error probability of rejecting the same fingerprint as a different fingerprint. FRR= (number of rejected fingerprints/total number of fingerprints investigated) × 100%.

False acceptance rate (FAR): refers to the error probability that different fingerprints are mistaken for the same fingerprint. FAR= (number of misjudged fingerprints/total number of fingerprints checked) × 100%.

For an existing system, by setting different system thresholds, we can see that these two indicators are interrelated, and FRR is inversely proportional to FAR. It is easy to understand that the stricter the investigation, the lower the possibility of misunderstanding, but the higher the possibility of refusing to know.

2.2 Test methods of false recognition rate and rejection rate

In order to test these two indicators, the method of cyclic testing is usually adopted [7]. That is to say, given a group of images, then combine them in turn, submit and compare them, count the total number of submissions and errors, and calculate the proportion of errors, that is, FRR and FAR. For the index of FAR=0.000 1%, not less than 1.4 15 different fingerprint images should be used for cyclic testing, and the total number of tests is 1.000405. If the error comparison is successful in the test, far =11.000. For FRR=0. 1%, at least 46 images belonging to the same fingerprint should be paired for testing, so the total number of times to submit the test is 1.035. If it is rejected by mistake, FRR =11.035. The more samples are used in the test, the more accurate the results will be. The fingerprint image as a test sample should meet the registrable conditions.

2.3 System parameters

Error Registration Rate (ERR): refers to the probability that fingerprints cannot be registered and processed on fingerprint devices. If the ERR is too high, it will seriously affect the application range of the device, usually less than 1%.

Login time: the time required for the fingerprint device to login the fingerprint. Usually, the time required for a single login is less than 2 s.

Comparison time: the time taken by the fingerprint device to compare two sets of fingerprint feature templates is usually less than1s.

Working temperature: the allowable temperature change range when the fingerprint meter works normally, generally 0 ~ 40℃.

Working humidity: the allowable relative humidity range of fingerprint equipment during normal operation, generally 30% ~ 95%.

3 the application of fingerprint identification technology

Fingerprint identification technology has matured and its application is becoming more and more common. In addition to criminal investigation, it is also widely used in civil applications, such as fingerprint access control system, fingerprint attendance system, bank fingerprint savings system, bank fingerprint safe deposit box, fingerprint medical insurance system, family planning fingerprint management system, children's pick-up fingerprint management system, fingerprint blood donation management system, securities trading fingerprint system, fingerprint gun management system, intelligent building fingerprint access control system, driver fingerprint management system and so on.

Fingerprint access control system and fingerprint attendance system are the earliest access control management systems, including intercom fingerprint access control, online fingerprint access control, offline fingerprint access control and so on. Press an individual's finger on the fingerprint collector at the entrance, and the system will compare the fingerprints registered in the fingerprint database (called registered). If it matches (that is, matches), it will be successful and the door will be opened automatically. If it doesn't match, it will display "unsuccessful" or "without this fingerprint" and the door won't open. In the fingerprint access control system, it can be a one-to-one comparison (one? Where to? A match), can also be a one-to-many comparison (a? Where to? The former can be a company or department, and the latter can be a confidential place such as family, bank business hall, vault, finance department and warehouse. In these applications, fingerprint identification system will replace or supplement many widely used photo and ID card systems.

Combining fingerprint identification technology with IC card is one of the most promising applications at present. This technology stores the fingerprint (encryption) of the owner of the card on the IC card, and adds a fingerprint identification system to the reader of the IC card. When the card reader reads the information on the card, it also reads the cardholder's fingerprint. Through comparison, it can be confirmed whether the cardholder is the real owner of the card, so as to carry out the next transaction. Fingerprint IC card can replace the existing ATM card to make anti-counterfeiting documents. ATM card holders don't have to use passwords, which avoids the difficulty for the elderly and children to remember passwords.

In recent years, the Internet has brought convenience and benefits to people, but there are also security problems. Fingerprint feature data can be transmitted and verified on the computer network by email or other transmission methods. Through fingerprint identification technology, only designated people can access relevant information, which can greatly improve the security of online information. A series of online business activities, such as online banking, online trade and e-commerce, are all protected by security.

The application of fingerprint social insurance system has played a very effective role in the accurate payment of pension. It avoids the problem that someone else uses a seal or a copy of the ID card to collect it, and the issuer cannot be sure that the person has died. Only by relying on my living fingerprint can the pension be paid accurately.

4 the reliability of fingerprint identification

Fingerprint identification technology is a mature biometric technology. Because everyone's skin lines, including fingerprints, patterns, breakpoints and intersections are different, they are unique and will remain unchanged for life. By comparing his fingerprints with pre-stored fingerprints, we can verify his true identity. Automatic fingerprint identification is a method of fingerprint identification by computer. It benefits from modern electronic integrated manufacturing technology and fast and reliable algorithm theory research. Although fingerprints are only a small part of human skin, the amount of data used for identification is quite large. Comparing these data is a fuzzy matching algorithm, which requires a lot of operations. Small fingerprint image reading equipment and faster computer produced by modern electronic integrated manufacturing technology provide the possibility for fingerprint comparison operation on microcomputer. In addition, the reliability of the matching algorithm is also improving. Therefore, fingerprint identification technology has been very simple and practical. Because the computer only involves some limited information when processing fingerprints and the matching algorithm is not very accurate, the accuracy of 100% cannot be guaranteed.

The recognition rate is an important index to measure the specific application of fingerprint identification system. It mainly includes rejection rate and misidentification rate, which are inversely proportional. Adjust these two values for different purposes. Although there are reliability problems in the fingerprint identification system, its security is much higher than the "user ID+ password" scheme with the same reliability level. Rejection rate is actually an important indicator of system availability. In the design of application system, we should weigh the ease of use and security. Usually, comparing two or more fingerprints can greatly improve the security of the system without losing usability.

References:

/journal/articleinfo.aspx？ art_id= 1555 12