Fingerprint identification
        Fingerprint refers to the uneven texture on the front side of the finger. This pattern contains a lot of feature information, such as pattern, pattern area, core point, triangle point and number of lines, and the detail features, endpoint, breakpoint, bifurcation point, triangle point, core point, etc. are called "feature points". The two fingerprints often have the same general characteristics, but their details are not the same.

Figure 1 fingerprint features
       Fingerprint recognition is the only way to confirm a person's identity by relying on the overall characteristics and details. Fingerprint recognition technology is currently the most mature and widely used biometric technology. Its main characteristics are uniqueness, invariance and portability.

Technical process
       As a popular biometric technology, fingerprint recognition has been widely concerned by the industry. There are various fingerprint identification devices on the market, but the main steps involved in the identification process are the process of fingerprint image acquisition, fingerprint image preprocessing, fingerprint image feature extraction, matching and recognition. . Fingerprint image preprocessing usually includes image enhancement, two valued and thinning.

       The acquisition of fingerprint image is an important part of the automatic fingerprint identification system. The image acquisition is the process of collecting the living fingerprint image by a special fingerprint collector. With the development of semiconductor technology, a new type of sensor, such as CMOS fingerprint sensor, thermosensitive sensor and ultrasonic sensor, has been developed in succession with the development of semiconductor technology. At this stage, the sensor technology is becoming more and more exquisite, and the acquisition performance is also improving.

       Preprocessing of fingerprint images: the acquired fingerprint images are usually accompanied by a variety of noise, part of which are caused by the acquisition instrument, and the other is caused by the state of the finger. The purpose of fingerprint image preprocessing is to remove the noise in the image, make the image clear and edge obvious, and turn it into a clear point line map, so as to extract the correct fingerprint features. Fingerprint image pre-processing plays an important role in the whole fingerprint recognition system. Its quality directly affects the effect of fingerprint identification.

       Image enhancement: the most important step in the process of fingerprint preprocessing is to enhance the fingerprint image. The purpose of image enhancement is to suppress noise while enhancing the contrast of the ridge line structure, to connect the broken ridges and to separate the ridges, to highlight certain information in an image at a specific requirement, and to weaken or reduce it at the same time. Remove some of the information that is not needed. Fingerprint image enhancement algorithm is mainly based on Gabor filtering enhancement method and low quality fingerprint enhancement algorithm based on Fourier filtering. Gabor filter enhancement is enhanced by directional field image, and is the most common one in fingerprint enhancement algorithm. The low quality fingerprint enhancement algorithm based on Fourier filter is used to transform the fingerprint image enhancement from the space domain to the frequency domain by Fourier transform, and then the fingerprint image is band-pass filtering and directional filtering in the frequency domain, thus the fingerprint image is enhanced.

       Two value: the ridge (ridge) part of the image is enhanced after the image is enhanced, but the strength of the ridge is not exactly the same, which shows the difference of the gray value. The purpose of the two value is to make the gray value of the ridges consistent and make the whole image simplified to two yuan information in the fingerprint recognition. On the one hand, the image information is compressed, the main information of the lines is preserved, the storage space is saved, and the other side can also remove a large number of adhesions for the extraction and matching of the fingerprint features. Prepare.

       Refinement: after the two value of the fingerprint image, the ridge still has a certain width, and fingerprint recognition is only interested in the trend of the ridge, and does not care about its thickness. The purpose of refinement is to delete the edge pixels of the fingerprint lines to make it only one pixel width, reduce the redundant information, highlight the main features of the fingerprint lines, and thus facilitate the extraction of the features behind the fingerprint. When thinning, we should ensure the connectivity of the ridge, the directionality and the feature points.

       Feature extraction of fingerprint images: there are two main feature extraction methods, one is to extract features from gray image, and the other is to extract features from two value images. The algorithm of extracting feature directly from gray image is to track the gray fingerprint lines, and find the location of the feature and the type of the judgment feature according to the tracking result. This method saves the process of complex fingerprint image preprocessing, but the algorithm of feature extraction is very complex, and the feature information (location, direction, etc.) is not accurate because of the influence of noise and other factors. The method of extracting features from two value images is simple. After a reliable two value image is refined, only a 3 * 3 template can be extracted from the endpoint and the bifurcation point.

       Matching and recognition is the last step in fingerprint identification system, and is also the most important basis for evaluating the performance of the whole fingerprint recognition system. Fingerprint matching is based on the extracted fingerprint features to determine whether the two fingerprints come from the same finger. The feature matching is mainly the matching of the detail features, and compares the details of the new input fingerprint with the details of the fingerprint stored in the fingerprint library to find out the most similar fingerprint as the output of the recognition, that is, the process of fingerprint verification and recognition, which is the ultimate aim of the fingerprint identification system. Because of the influence of various factors, the feature templates obtained by the two input of the same fingerprint are very different. Therefore, as long as the details of the input fingerprint are similar to those stored in the template, these two fingerprints are matched.

Chart 2 fingerprint identification system flow chart
       In fingerprint identification technology, fingerprint matching algorithm that can match and recognize feature values can be divided into: 
       One to one validation and one pair of N recognition.
       One to one validation is a validation process. Using the ID of the personnel, the fingerprint features were extracted first from the fingerprint library, and then compared with the fingerprint features extracted from the fingerprint, to prove that the person is the person to be identified.
       A pair of N recognition is a process of identification. The fingerprint characteristic value is extracted from the fingerprint, and the fingerprint characteristic value is compared with the fingerprint characteristic value stored in the fingerprint, and the fingerprint matching with the field fingerprint is found and the identity of the personnel is identified.

Technical advantage
    Tiancheng Industry is one of the earliest units engaged in the research and development of fingerprint technology in China, its independent research and development of fingerprint identification core algorithm FPI, research and development cycle of more than ten years, after more than 12 years of stable application, technology is very mature, technical indicators in the leading position at home and abroad. Fingerprint related products are the core technology of FPI fingerprint recognition algorithm, including fingerprint authentication instrument, SmartFPI fingerprint authentication platform, biometric unified identity authentication platform SmartBIOS, and many fingerprint management systems. Tiancheng Industry took the lead in putting forward the concept of "fingerprint payment" in China. In 2008, the first fingerprint bank in China was introduced, and fingerprint technology was applied to the external customers of the bank. In a sense, it led the application process of fingerprint technology in domestic banks.

       After more than ten years of accumulation and precipitation, all the products of the company's hardware and software have been developed and designed by themselves. They have all their own intellectual property rights. At present, 27 patents have been applied successfully in the field of Tiancheng Industry. "To solve the low quality fingerprint of fingerprints" has won the two prize of the Beijing science and technology award. At the same time, Tiancheng is invited to join the National Standard Technical Committee on Information Technology (SAC/TC28) as the main contributor to the national standards for biometrics, and has been involved in the formulation of the BioAPI national standard 1.1 edition of the relevant standards for fingerprint identification.

Application field
       With the progress of science and technology, the innovation of sensor technology and the development of fingerprint identification technology provide wider space for the application of fingerprint identification technology. In recent years, the system of fingerprint verification and recognition has been used in large scale. Fingerprint recognition has been applied to all aspects of life:
Computer network security: fingerprint recognition and traditional identity authentication are combined to prevent the password from being forgotten or stolen by others.
Financial institutions: banks and other institutions for staff identity authentication, ATM customer identity authentication, mobile payment security certification;
Government organs and enterprises and institutions are mainly used to authorize and authenticate the managers, check the attendance, verify the identity of the staff in the security area, and control the security of the database access.
Education, medical and judicial institutions: the school through entrance control control of the identity of students and staff; hospital and other security authentication of private information, so that the end users can safely and conveniently inquire and verify the relevant procedures, patients and donors to identify, prevent fraudulent donations and so on; public security and judicial organs use fingerprints. To verify the solution and identify the dangerous criminals quickly and effectively so as to ensure the safety of the citizens.
Border control, identity authentication on passport and visa of airport, port entry and exit personnel, and blacklist screening in border inspection.